Linux DeepSpeed - Text-generation-webui
@@ -469,8 +470,7 @@ Finetuning note - The **Nvidia CUDA Development Toolkit v11.8** is essential for 1. **Nvidia CUDA Development Toolkit Installation**: - The toolkit is crucial for DeepSpeed to compile/build for your version of Linux and requires around 3GB's of disk space. - - Install using your package manager e.g. **CUDA Toolkit 11.8** or download directly from [Nvidia CUDA Toolkit Archive](https://developer.nvidia.com/cuda-toolkit-archive) (choose 11.8 or 12.1 for Linux). - - If you want to use Finetuning with AllTalk, version **11.8** is required; otherwise, any version from 11.8 upwards will suffice. + - Install using your package manager **(Recommended)** e.g. **CUDA Toolkit 11.8** or download directly from [Nvidia CUDA Toolkit Archive](https://developer.nvidia.com/cuda-toolkit-archive) (choose 11.8 or 12.1 for Linux). 2. **Open a Terminal Console**: - After Nvidia CUDA Development Toolkit installation, access your terminal console. @@ -517,8 +517,7 @@ Finetuning note - The **Nvidia CUDA Development Toolkit v11.8** is essential for 1. **Nvidia CUDA Development Toolkit Installation**: - The toolkit is crucial for DeepSpeed to compile/build for your version of Linux and requires around 3GB's of disk space. - - Install using your package manager e.g. **CUDA Toolkit 11.8** or download directly from [Nvidia CUDA Toolkit Archive](https://developer.nvidia.com/cuda-toolkit-archive) (choose 11.8 or 12.1 for Linux). - - If you want to use Finetuning with AllTalk, version **11.8** is required; otherwise, any version from 11.8 upwards will suffice. + - Install using your package manager **(Recommended)** e.g. **CUDA Toolkit 11.8** or download directly from [Nvidia CUDA Toolkit Archive](https://developer.nvidia.com/cuda-toolkit-archive) (choose 11.8 or 12.1 for Linux). 2. **Open a Terminal Console**: - After Nvidia CUDA Development Toolkit installation, access your terminal console. @@ -840,7 +839,7 @@ This is a long standing issue with Firefox and one I am unable to resolve. The s ### Application Specific Issues🟨 SillyTavern - I changed my IP address and now SillyTavern wont connect with AllTalk
+
🟨 SillyTavern - I changed my IP address and now SillyTavern wont connect with AllTalk
SillyTavern checks the IP address when loading extensions, saving the IP to its configuration only if the check succeeds. For whatever reason, SillyTavern's checks dont always allow changing its IP address a second time.
To manually change the IP address: @@ -891,7 +890,6 @@ If you wish to try out the XTTS version 2.0.3 model and see if it works better, If you have a voice that the model doesnt quite reproduce correctly, or indeed you just want to improve the reproduced voice, then finetuning is a way to train your "XTTSv2 local" model **(stored in `/alltalk_tts/models/xxxxx/`)** on a specific voice. For this you will need: - An Nvidia graphics card. (Please see the help section [note](https://github.com/erew123/alltalk_tts/edit/main/README.md#performance-and-compatibility-issues) if you have multiple Nvidia GPU's). -- To install a few portions of the Nvidia CUDA Development Toolkit **version 11.8**. Explicitly version 11.8. - 18GB of disk space free (most of this is used temporarily) - At least 2 minutes of good quality speech from your chosen speaker in mp3, wav or flacc format, in one or more files (have tested as far as 20 minutes worth of audio). - As a side note, many people seem to think that the Whisper v2 model (used on Step 1) is giving better results at generating training datasets, so you may prefer to try that, as opposed to the Whisper 3 model. @@ -906,15 +904,25 @@ I would suggest that if its in an interview format, you cut out the interviewer Yes you can. You would do these as multiple finetuning's, but its absolutely possible and fine to do. Finetuning the XTTS model does not restrict it to only being able to reproduce that 1x voice you trained it on. Finetuning is generally nuding the model in a direction to learn the ability to sound a bit more like a voice its not heard before. #### ⚫ A note about anonymous training Telemetry information & disabling it -Portions of Coqui's TTS trainer scripts gather anonymous training information which you can disable. Their statement on this is listed [here](https://github.com/coqui-ai/Trainer?tab=readme-ov-file#anonymized-telemetry). Although I have tried to pass `TRAINER_TELEMETRY=0` through AllTalk, it appears you will have to set this manually if you wish to finetune on a non-internet enabled computer or disable the anonymous data being sent. You can do this by: +Portions of Coqui's TTS trainer scripts gather anonymous training information which you can disable. Their statement on this is listed [here](https://github.com/coqui-ai/Trainer?tab=readme-ov-file#anonymized-telemetry). If you start AllTalk Finetuning with `start_finetuning.bat` or `./start_finetuning.sh` telemetry will be disabled. If you manually want to disable it, please expand the below: + +
Manually disable telemetry
+ +Before starting finetuning, run the following in your terminal/command prompt: - On Windows by typing `set TRAINER_TELEMETRY=0` - On Linux & Mac by typing `export TRAINER_TELEMETRY=0` -Before you start `finetune.py`. You will now be able to finetune offline and no anonymous training data will be sent. +Before you start `finetune.py`. You will now be able to finetune offline and no anonymous training data will be sent. +
Legacy manual instructions for installing Nvidia CUDA Development Toolkit 11.8
- To perform fine-tuning, a specific portion of the **Nvidia CUDA Development Toolkit v11.8** must be installed. This is crucial for step 1 of fine-tuning. The objective is to minimize the installation footprint by installing only the essential components. - The **Nvidia CUDA Development Toolkit v11.8** operates independently from your graphics card drivers and the CUDA version utilized by your Python environment. - This installation process aims to keep the download and install size as minimal as possible, however a full install of the tookit requires 3GB's of disk space. @@ -954,11 +962,13 @@ Before you start `finetune.py`. You will now be able to finetune offline and no 5. **Troubleshooting**: - If the correct version isn't reported, recheck your environment path settings for accuracy and potential conflicts with other CUDA versions. +
+ #### ⚫ Starting Fine-tuning **NOTE:** Ensure AllTalk has been launched at least once after any updates to download necessary files for fine-tuning. @@ -972,6 +982,13 @@ Before you start `finetune.py`. You will now be able to finetune offline and no Depending on your setup (Text-generation-webui or Standalone AllTalk), the steps to start the Python environment vary: +- **For Standalone AllTalk Users**: + - Navigate to the `alltalk_tts` folder: + - `cd alltalk_tts` + - Start the Python environment: + - Windows: `start_finetune.bat` + - Linux: `./start_finetune.sh` + - **For Text-generation-webui Users**: - Navigate to the Text-generation-webui directory: - `cd text-generation-webui` @@ -982,29 +999,19 @@ Depending on your setup (Text-generation-webui or Standalone AllTalk), the steps - WSL (Windows Subsystem for Linux): `cmd_wsl.bat` - Move into the AllTalk directory: - `cd extensions/alltalk_tts` + - **Linux** users only need to run this command: + ``` + export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'` + ``` + - Start the fine-tuning process with the command: + - `python finetune.py`
-- **For Standalone AllTalk Users**: - - Navigate to the `alltalk_tts` folder: - - `cd alltalk_tts` - - Start the Python environment: - - Windows: `start_environment.bat` - - Linux: `./start_environment.sh`
- > If you're unfamiliar with Python environments and wish to learn more, consider reviewing **Understanding Python Environments Simplified** in the Help section. -3. **Install Fine-tuning Requirements**: - - If you haven't already, install the fine-tuning requirements: - - `pip install -r requirements_finetune.txt` - -4. **Initiate Fine-tuning**: - - Start the fine-tuning process with the command: - - `python finetune.py` - - From now on you will only need to start your Python environment, potentially set the CUDA path and run the above command when you want to use Finetuning. - -5. **Pre-Flight Checklist**: +3. **Pre-Flight Checklist**: - Go through the pre-flight checklist to ensure readiness. Address any issues flagged as "Fail". -6. **Post Fine-tuning Actions**: +4. **Post Fine-tuning Actions**: - Upon completing fine-tuning, the final tab will guide you on managing your files and relocating your newly trained model to the appropriate directory. These steps guide you through the initial preparations, starting the Python environment based on your setup, and the fine-tuning process itself. Ensure all prerequisites are met to facilitate a smooth fine-tuning experience. @@ -1059,6 +1066,13 @@ This will vary by system for a multitude of reasons, however, while generating a - **Stop Playback:** This will stop the current audio playing back. It does not stop the text from being generated however. Once you have sent text off to be generated, either as a stream or wav file generation, the TTS server will remain busy until this process has competed. As such, think carefully as to how much you want to send to the server. If you are generating wav files and populating the queue, you can generate one lot of text to speech, then input your next lot of text and it will continue adding to the list. +#### ⬜ Customization and Preferences +- **Character Voice:** Choose the voice that will read your text. +- **Language:** Select the language of your text. +- **Chunk Sizes:** Decide the size of text chunks for generation. Smaller sizes are recommended for better TTS quality. +#### ⬜ Interface and Accessibility +- **Dark/Light Mode:** Switch between themes for your visual comfort. +- **Word Count and Generation Queue:** Keep track of the word count and the generation progress. #### ⬜ TTS Generation Modes - **Wav Chunks:** Perfect for creating audio books, or anything you want to keep long term. Breaks down your text into manageable wav files and queues them up. Generation begins automatically, and playback will start after a few chunks have been prepared ahead. You can set the volume to 0 if you don’t want to hear playback. With Wav chunks, you can edit and/or regenerate portions of the TTS as needed. - **Streaming:** For immediate playback without the ability to save. Ideal for on-the-fly speech generation and listening. This will not generate wav files and it will play back through your browser. You cannot stop the server generating the TTS once it has been sent.
@@ -1071,13 +1085,23 @@ For generating **large amounts of TTS**, it's recommended to select the **No Pla - **Export/Import List:** Save your TTS list as a JSON file or import one. Note: Existing wav files are needed for playback. Exporting is handy if you want to take your files away into another program and have a list of which wav is which, or if you keep your audio files, but want to come back at a later date, edit one or two lines, regenerate the speech and re-combine the wav’s into one new long wav. #### ⬜ Exporting Your Audio - **Export to WAV:** Combine all generated TTS from the list, into one single WAV file for easy download and distribution. Its always recommended to export your list to a JSON before exporting, so that you have a backup, should something go wrong. You can simply re-import the list and try exporting again.
When exporting, there is a file size limit of 1GB and as such you have the option to choose how many files to include in each block of audio exported. 600 is just on the limit of 1GB, depending on the average file size, so 500 or less is a good amount to work with. You can combine the generated files after if you wish, in Audacity or similar.
Additionally, lower export batches will lower the memory requirements, so if your system is low on memory (maybe 8 or 16GB system), you can use smaller export batches to keep the memory requirement down. -#### ⬜ Customization and Preferences -- **Character Voice:** Choose the voice that will read your text. -- **Language:** Select the language of your text. -- **Chunk Sizes:** Decide the size of text chunks for generation. Smaller sizes are recommended for better TTS quality. -#### ⬜ Interface and Accessibility -- **Dark/Light Mode:** Switch between themes for your visual comfort. -- **Word Count and Generation Queue:** Keep track of the word count and the generation progress. +#### ⬜ Exporting Subtitles (SRT file) +- **Export SRT:** This will scan through all wav files in your list and generate a subtitles file that will match your exported wav file. +#### ⬜ Analyzing generated TTS for errors +- **Analyze TTS:** This will scan through all wav files comparing each ID's orignal text with the TTS generated for that ID and then flag up inconsistences. Its important to understand this is a **best effort** process and **not 100% perfect**, for example:
+ - Your text may have the word `their` and the automated routine that listens to your generated TTS interprets the word as `there`, aka a spelling difference. + - Your text may have `Examples are:` (note the colon) and the automated routine that listens to your generated TTS interprets the word as "Examples are` (note no colon as you cannot sound out a colon in TTS), aka a punctuation difference. + - Your text may have `There are 100 items` and the automated routine that listens to your generated TTS interprets the word as `There are one hundred items`, aka numbers vs the number written out in words. + - There will be other examples such as double quotes. As I say, please remember this is a **best effort** to help you identify issues.
+ +As such, there is a `% Accuracy` setting. This uses a couple of methods to try find things that are similar e.g. taking the `their` and `there` example from above, it would identify that they both sound the same, so even if the text says `their` and the AI listening to the generated TTS interprets the word as `there`, it will realise that both sound the same/are similar so there is no need to flag that as an error. However, there are limits to this and some things may slip through or get picked up when you would prefer them not to be flagged. + +The higher the accuracy you choose, the more things it will flag up, however you may get more unwanted detections. The lower the less detections. Based on my few tests, accuracy settings between 96 to 98 seem to generally give the best results. However, I would highly recommend you test out a small 10-20 line text and test out the **Analyze TTS** button to get a feel for how it responds to different settings, as well as things it flags up. + +You will be able to see the ID's and Text (orignal and as interpreted) by looking at the terminal/command prompt window. + +The Analyze TTS feature uses the Whisper Larger-v2 AI engine, which will download on first use if necessary. This will require about 2.5GB's of disk space and could take a few minutes to download, depending on your internet connection. + #### ⬜ Tricks to get the model to say things correctly Sometimes the AI model won’t say something the way that you want it to. It could be because it’s a new word, an acronym or just something it’s not good at for whatever reason. There are some tricks you can use to improve the chances of it saying something correctly. diff --git a/atsetup.bat b/atsetup.bat index be6167d..b4cfbfc 100644 --- a/atsetup.bat +++ b/atsetup.bat @@ -61,11 +61,10 @@ echo environment. If you have NOT done this, please run %L_GREEN%cmd_windows. echo in the %L_GREEN%text-generation-webui%RESET% folder and then re-run this script. echo. echo BASE REQUIREMENTS -echo 1) Install the requirements for a %L_GREEN%Nvidia machine%RESET%. -echo 2) Install the requirements for a %L_GREEN%AMD or MAC machine%RESET%. +echo 1) Apply/Re-Apply the requirements for %L_GREEN%Text-generation-webui%RESET%. echo. echo OPTIONAL -echo 3) Install the Finetuning requirements. +echo 2) Git Pull the latest AllTalk updates from Github echo. echo DEEPSPEED FOR %L_YELLOW=%PyTorch 2.1.x%RESET% echo 4) Install DeepSpeed v11.2 for CUDA %L_GREEN%11.8%RESET% and Python-3.11.x and %L_YELLOW%PyTorch 2.1.x%RESET%. @@ -84,8 +83,7 @@ echo 9) %L_RED%Exit/Quit%RESET% echo. set /p WebUIOption="Enter your choice: " if "%WebUIOption%"=="1" goto InstallNvidiaTextGen -if "%WebUIOption%"=="2" goto InstallOtherTextGen -if "%WebUIOption%"=="3" goto InstallFinetuneTextGen +if "%WebUIOption%"=="2" goto TGGitpull if "%WebUIOption%"=="4" goto InstallDeepSpeed118TextGen if "%WebUIOption%"=="5" goto InstallDeepSpeed121TextGen if "%WebUIOption%"=="6" goto InstallDeepSpeed121TextGenPytorch221 @@ -105,27 +103,31 @@ echo BASE REQUIREMENTS echo 1) Install AllTalk as a Standalone Application echo. echo OPTIONAL -echo 2) Delete AllTalk's custom Python environment -echo 3) Install the Finetuning requirements +echo 2) Git Pull the latest AllTalk updates from Github +echo 3) Re-Apply/Update the requirements file +echo 4) Delete AllTalk's custom Python environment +echo 5) Purge the PIP cache echo. echo. OTHER -echo 4) Generate a diagnostics file +echo 8) Generate a diagnostics file echo. echo 9) %L_RED%Exit/Quit%RESET% echo. set /p StandaloneOption="Enter your choice: " if "%StandaloneOption%"=="1" goto InstallCustomStandalone -if "%StandaloneOption%"=="2" goto DeleteCustomStandalone -if "%StandaloneOption%"=="3" goto InstallFinetuneStandalone -if "%StandaloneOption%"=="4" goto GenerateDiagsStandalone +if "%StandaloneOption%"=="2" goto STGitpull +if "%StandaloneOption%"=="3" goto STReapplyrequirements +if "%StandaloneOption%"=="4" goto STDeleteCustomStandalone +if "%StandaloneOption%"=="5" goto STPurgepipcache +if "%StandaloneOption%"=="8" goto GenerateDiagsStandalone if "%StandaloneOption%"=="9" goto End goto StandaloneMenu :InstallNvidiaTextGen -pip install -r requirements_nvidia.txt +pip install -r system\requirements\requirements_textgen.txt if %ERRORLEVEL% neq 0 ( echo. - echo There was an error installing the Nvidia requirements. + echo There was an error installing the requirements. echo Press any key to return to the menu. echo. pause @@ -133,46 +135,28 @@ if %ERRORLEVEL% neq 0 ( ) Echo. echo. -Echo Nvidia machine requirements installed successfully. +Echo Requirements installed successfully. Echo. pause goto WebUIMenu -:InstallOtherTextGen -pip install -r requirements_other.txt +:TGGitpull +git pull if %ERRORLEVEL% neq 0 ( - echo. - echo There was an error installing the Other requirements. + echo. + echo There was an error pulling from Github. echo Press any key to return to the menu. echo. pause goto WebUIMenu ) Echo. -echo. -Echo Other machine requirements installed successfully. +echo AllTalk Updated from Github. Please re-apply +echo the latest requirements file. Option 1 Echo. pause goto WebUIMenu -:InstallFinetuneTextGen -pip install -r requirements_finetune.txt -if %ERRORLEVEL% neq 0 ( - echo. - echo There was an error installing the Finetune requirements. - echo Press any key to return to the menu. - echo. - pause - goto WebUIMenu -) -Echo. -Echo Finetune requirements installed successfully. Please ensure you have installed CUDA 11.8. -Echo Finetune needs access to %L_GREEN%cublas64_11.dll%RESET% and will error if this is not setup correctly. -Echo Please see here for details %L_GREEN%https://github.com/erew123/alltalk_tts#-finetuning-a-model%RESET% -Echo. -pause -goto WebUIMenu - :InstallDeepSpeed118TextGen echo Downloading DeepSpeed... curl -LO https://github.com/erew123/alltalk_tts/releases/download/deepspeed/deepspeed-0.11.2+cuda118-cp311-cp311-win_amd64.whl @@ -394,49 +378,20 @@ if not exist "%INSTALL_ENV_DIR%\python.exe" ( echo. && echo Conda environment is @rem activate installer env call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" || ( echo. && echo Miniconda hook not found. && goto end ) -@rem Ask user for the type of requirements to install echo. -echo %L_BLUE%Choose the type of requirements to install%RESET% +echo Downloading and installing PyTorch. This step can take a long time +echo depending on your internet connection and hard drive speed. Please +echo be patient. echo. -echo 1. Nvidia graphics card machines -echo 2. Other machines (mac, amd, etc) +pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121 +echo Installing other requirements. echo. -set /p UserChoice="Enter your choice (1 or 2): " +pip install -r system\requirements\requirements_standalone.txt +curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl +echo Installing DeepSpeed... +pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl +del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl -@rem Install requirements based on user choice -if "%UserChoice%" == "1" ( - echo. - echo Installing Nvidia requirements... - echo Downloading Pytorch with CUDA.. - pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --upgrade --force-reinstall --index-url https://download.pytorch.org/whl/cu121 - pip install numpy==1.24.4 - pip install typing-extensions>=4.8.0 - pip install networkx<3.0.0,>=2.5.0 - pip install fsspec>=2023.6.0 - pip install soundfile==0.12.1 - pip install uvicorn==0.24.0.post1 - pip install transformers>=4.37.1 - pip install TTS==0.22.0 - pip install fastapi==0.104.1 - pip install Jinja2==3.1.2 - pip install requests==2.31.0 - pip install tqdm==4.66.1 - pip install importlib-metadata==4.8.1 - pip install packaging==23.2 - pip install pydantic==1.10.13 - pip install sounddevice==0.4.6 - pip install python-multipart==0.0.6 - pip install cutlet>=0.3.0 - pip install unidic-lite>=1.0.8 - echo Downloading DeepSpeed... - curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl - echo Installing DeepSpeed... - pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl - del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl -) else ( - echo Installing requirements for other machines... - pip install -r requirements_other.txt -) @rem Create start_environment.bat to run AllTalk environment echo @echo off > start_environment.bat @@ -451,14 +406,22 @@ echo set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda >> start_alltalk.bat echo set INSTALL_ENV_DIR=%cd%\alltalk_environment\env >> start_alltalk.bat echo call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" >> start_alltalk.bat echo call python script.py >> start_alltalk.bat +@rem Create start_finetune.bat to run AllTalk +echo @echo off > start_finetune.bat +echo cd /D "%~dp0" >> start_finetune.bat +echo set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda >> start_finetune.bat +echo set INSTALL_ENV_DIR=%cd%\alltalk_environment\env >> start_finetune.bat +echo call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" >> start_finetune.bat +echo call python finetune.py >> start_finetune.bat Echo. -Echo start_alltalk.bat has been created. -Echo You can now start AllTalk with %L_YELLOW%start_alltalk.bat%RESET% +Echo Run %L_YELLOW%start_alltalk.bat%RESET% to start AllTalk. +Echo Run %L_YELLOW%start_finetune.bat%RESET% to start Finetuning. +Echo Run %L_YELLOW%start_environment.bat%RESET% to start the AllTalk Python environment. Echo. pause goto StandaloneMenu -:DeleteCustomStandalone +:STDeleteCustomStandalone @rem Check if the alltalk_environment directory exists if not exist "%cd%\alltalk_environment\" ( echo. @@ -492,47 +455,6 @@ Echo. pause goto StandaloneMenu -:InstallFinetuneStandalone -cd /D "%~dp0" -set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda -set INSTALL_ENV_DIR=%cd%\alltalk_environment\env -@rem Check if the Conda environment exists -if not exist "%INSTALL_ENV_DIR%\python.exe" ( - echo. - echo The Conda environment at "%INSTALL_ENV_DIR%" does not exist. - echo Please install the environment before proceeding. - echo. - pause - goto StandaloneMenu -) -@rem Attempt to activate the Conda environment -call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" -if errorlevel 1 ( - echo. - echo Failed to activate the Conda environment. - echo Please check your installation and try again. - echo. - pause - goto StandaloneMenu -) -@rem Proceed with installing requirements -pip install -r requirements_finetune.txt -if %ERRORLEVEL% neq 0 ( - echo. - echo There was an error installing the Finetune requirements. - echo Press any key to return to the menu. - echo. - pause - goto StandaloneMenu -) -Echo. -Echo Finetune requirements installed successfully. Please ensure you have installed CUDA 11.8. -Echo Finetune needs access to %L_GREEN%cublas64_11.dll%RESET% and will error if this is not setup correctly. -Echo Please see here for details %L_GREEN%https://github.com/erew123/alltalk_tts#-finetuning-a-model%RESET% -Echo. -pause -goto StandaloneMenu - :GenerateDiagsStandalone cd /D "%~dp0" set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda @@ -573,6 +495,144 @@ Echo. pause goto StandaloneMenu +:STReapplyrequirements +cd /D "%~dp0" +set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda +set INSTALL_ENV_DIR=%cd%\alltalk_environment\env +@rem Check if the Conda environment exists +if not exist "%INSTALL_ENV_DIR%\python.exe" ( + echo. + echo The Conda environment at "%INSTALL_ENV_DIR%" does not exist. + echo Please install the environment before proceeding. + echo. + pause + goto StandaloneMenu +) +@rem Attempt to activate the Conda environment +call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" +if errorlevel 1 ( + echo. + echo Failed to activate the Conda environment. + echo Please check your installation and try again. + echo. + pause + goto StandaloneMenu +) +@rem Run Reapply requirements +echo. +echo Downloading and installing PyTorch. This step can take a long time +echo depending on your internet connection and hard drive speed. Please +echo be patient. +echo. +pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121 +echo Installing other requirements. +echo. +pip install -r system\requirements\requirements_standalone.txt +curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl +echo Installing DeepSpeed... +pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl +del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl +if %ERRORLEVEL% neq 0 ( + echo. + echo There was an error. + echo Press any key to return to the menu. + echo. + pause + goto StandaloneMenu +) +Echo. +Echo. +Echo Requirements have been re-applied/updated. +Echo. +pause +goto StandaloneMenu + +:STPurgepipcache +cd /D "%~dp0" +set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda +set INSTALL_ENV_DIR=%cd%\alltalk_environment\env +@rem Check if the Conda environment exists +if not exist "%INSTALL_ENV_DIR%\python.exe" ( + echo. + echo The Conda environment at "%INSTALL_ENV_DIR%" does not exist. + echo Please install the environment before proceeding. + echo. + pause + goto StandaloneMenu +) +@rem Attempt to activate the Conda environment +call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" +if errorlevel 1 ( + echo. + echo Failed to activate the Conda environment. + echo Please check your installation and try again. + echo. + pause + goto StandaloneMenu +) +@rem Clear the PIP cache +echo. +echo Purging the PIP cache of downloaded files. +echo. +pip cache purge +if %ERRORLEVEL% neq 0 ( + echo. + echo There was an error. + echo Press any key to return to the menu. + echo. + pause + goto StandaloneMenu +) +Echo. +Echo The PIP cache has been purged. +Echo. +pause +goto StandaloneMenu + +:STGitpull +cd /D "%~dp0" +set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda +set INSTALL_ENV_DIR=%cd%\alltalk_environment\env +@rem Check if the Conda environment exists +if not exist "%INSTALL_ENV_DIR%\python.exe" ( + echo. + echo The Conda environment at "%INSTALL_ENV_DIR%" does not exist. + echo Please install the environment before proceeding. + echo. + pause + goto StandaloneMenu +) +@rem Attempt to activate the Conda environment +call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" +if errorlevel 1 ( + echo. + echo Failed to activate the Conda environment. + echo Please check your installation and try again. + echo. + pause + goto StandaloneMenu +) +@rem Pull from Github +echo. +echo Pulling the latest updates. Please re-apply +echo the latest requirements file. Option 3 +echo. +git pull +if %ERRORLEVEL% neq 0 ( + echo. + echo There was an error pulling from Github. + echo Press any key to return to the menu. + echo. + pause + goto StandaloneMenu +) +Echo. +echo AllTalk Updated from Github. Please re-apply +echo the latest requirements file. Option 3 +Echo. +pause +goto StandaloneMenu + :End echo Exiting AllTalk Setup Utility... exit /b diff --git a/atsetup.sh b/atsetup.sh index 7bd913f..61ff58a 100644 --- a/atsetup.sh +++ b/atsetup.sh @@ -64,11 +64,12 @@ webui_menu() { echo -e " in the ${L_GREEN}text-generation-webui${NC} folder and then re-run this script." echo echo " BASE REQUIREMENTS" - echo -e " 1) Install the AllTalk requirements for an ${L_GREEN}Nvidia GPU machine.${NC}" - echo -e " 2) Install the AllTalk requirements for an ${L_GREEN}AMD or MAC machine.${NC}" + echo -e " 1) Apply/Re-Apply the requirements for an ${L_GREEN}Text-generation-webui${NC}" echo echo " OPTIONAL" - echo " 3) Install the Finetuning requirements." + echo " 2) Git Pull the latest AllTalk updates from Github" + echo + echo " DEEPSPEED" echo " 4) Install DeepSpeed." echo " 5) Uninstall DeepSpeed." echo @@ -81,8 +82,7 @@ webui_menu() { case $webui_option in 1) install_nvidia_textgen ;; - 2) install_other_textgen ;; - 3) install_finetune_textgen ;; + 2) tg_gitpull ;; 4) install_deepspeed ;; 5) uninstall_deepspeed ;; 6) generate_diagnostics_textgen ;; @@ -93,52 +93,40 @@ webui_menu() { } install_nvidia_textgen() { - local requirements_file="requirements_nvidia.txt" - echo " Installing Finetune requirements from $requirements_file..." + local requirements_file="system/requirements/requirements_textgen.txt" + echo " Installing Requirements from $requirements_file..." if ! pip install -r "$requirements_file"; then - echo " There was an error installing the Nvidia requirements." + echo + echo " There was an error pulling from Github." echo " Please check the output for details." read -p " Press any key to return to the menu. " -n 1 echo return fi - echo " Nvidia requirements installed successfully." + echo + echo " Requirements installed successfully." + echo + echo -e " To install ${L_YELLOW}DeepSpeed${NC} on Linux, there are additional" + echo -e " steps required. Please see the Github or documentation on DeepSeed." + echo read -p " Press any key to continue. " -n 1 echo } -install_other_textgen() { - local requirements_file="requirements_other.txt" - echo " Installing Other machine requirements from $requirements_file..." - if ! pip install -r "$requirements_file"; then - echo " There was an error installing the Other requirements." - echo " Please check the output for details." - read -p " Press any key to return to the menu. " -n 1 - echo - return - fi - echo " Other machine requirements installed successfully." - read -p " Press any key to continue. " -n 1 +tg_gitpull() { echo -} - -install_finetune_textgen() { - local requirements_file="requirements_finetune.txt" - echo " Installing Finetune requirements from $requirements_file..." - if ! pip install -r "$requirements_file"; then - echo " There was an error installing the Finetune requirements." + if ! git pull; then + echo + echo " There was an error installing the requirements." echo " Please check the output for details." + echo read -p " Press any key to return to the menu. " -n 1 echo return fi echo - echo " Finetune base requirements installed successfully." - echo - echo -e " Finetuning requires access to ${GREEN}Cublas${NC} from the ${GREEN}Nvidia CUDA Toolkit${NC}, so will need installing." - echo - echo -e " Instructions can be found here:" - echo -e " ${GREEN}https://github.com/erew123/alltalk_tts?#-finetuning-a-model${NC}" + echo " AllTalk Updated from Github. Please re-apply" + echo " the latest requirements file. (Option 1)" echo read -p " Press any key to continue. " -n 1 echo @@ -185,12 +173,16 @@ uninstall_deepspeed() { echo "Uninstalling DeepSpeed..." pip uninstall -y deepspeed if [ $? -ne 0 ]; then + echo echo " There was an error uninstalling DeepSpeed." + echo echo " Press any key to return to the menu." read -n 1 return fi + echo echo " DeepSpeed uninstalled successfully." + echo echo " Press any key to continue." read -n 1 } @@ -198,13 +190,17 @@ uninstall_deepspeed() { generate_diagnostics_textgen() { # Run diagnostics if ! python diagnostics.py; then + echo echo " There was an error running diagnostics." + echo read -p " Press any key to return to the menu. " -n 1 echo return fi - - echo " Diagnostics completed successfully." + echo + echo " Diagnostics log file generated successfully." + echo " Please see diagnostics.log" + echo read -p " Press any key to continue. " -n 1 echo } @@ -220,12 +216,16 @@ standalone_menu() { echo " 1) Install AllTalk as a Standalone Application" echo echo " OPTIONAL" - echo " 2) Delete AllTalk's custom Python environment" - echo " 3) Install the Finetuning requirements" - echo " 4) Install DeepSpeed." + echo " 2) Git Pull the latest AllTalk updates from Github" + echo " 3) Re-Apply/Update the requirements file" + echo " 4) Delete AllTalk's custom Python environment" + echo " 5) Purge the PIP cache" + echo + echo " DEEPSPEED" + echo " 6) DeepSpeed Instructions/Install" echo echo " OTHER" - echo " 5) Generate a diagnostics file" + echo " 8) Generate a diagnostics file" echo echo -e " 9)${L_RED} Exit/Quit${NC}" echo @@ -233,16 +233,19 @@ standalone_menu() { case $standalone_option in 1) install_custom_standalone ;; - 2) delete_custom_standalone ;; - 3) install_finetune_standalone ;; - 4) install_deepspeed ;; - 5) generate_diagnostics_standalone ;; + 2) gitpull_standalone ;; + 3) reapply_standalone ;; + 4) delete_custom_standalone ;; + 5) pippurge_standalone ;; + 6) install_deepspeed ;; + 8) generate_diagnostics_standalone ;; 9) exit 0 ;; *) echo "Invalid option"; sleep 2 ;; esac done } + install_custom_standalone() { cd "$(dirname "${BASH_SOURCE[0]}")" @@ -284,39 +287,15 @@ install_custom_standalone() { # Activate the environment and install requirements source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh" conda activate "${INSTALL_ENV_DIR}" - echo - echo -e " ${L_BLUE}Choose the type of requirements to install${NC}" + echo " Downloading and installing PyTorch. This step can take a long time" + echo " depending on your internet connection and hard drive speed. Please" + echo " be patient." + pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121 echo - echo " 1. Nvidia graphics card machines" - echo " 2. Other machines (mac, amd, etc)" + echo " Installing additional requirements." echo - read -p " Enter your choice (1 or 2): " user_choice - - if [ "$user_choice" == "1" ]; then - pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --upgrade --force-reinstall --index-url https://download.pytorch.org/whl/cu121 - pip install numpy==1.24.4 - pip install typing-extensions>=4.8.0 - pip install networkx<3.0.0,>=2.5.0 - pip install fsspec>=2023.6.0 - pip install soundfile==0.12.1 - pip install uvicorn==0.24.0.post1 - pip install transformers>=4.37.1 - pip install TTS==0.22.0 - pip install fastapi==0.104.1 - pip install Jinja2==3.1.2 - pip install requests==2.31.0 - pip install tqdm==4.66.1 - pip install importlib-metadata==4.8.1 - pip install packaging==23.2 - pip install pydantic==1.10.13 - pip install sounddevice==0.4.6 - pip install python-multipart==0.0.6 - pip install cutlet>=0.3.0 - pip install unidic-lite>=1.0.8 - else - pip install -r requirements_other.txt - fi + pip install -r system/requirements/requirements_standalone.txt # Create start_environment.sh to run AllTalk cat << EOF > start_environment.sh #!/bin/bash @@ -333,27 +312,42 @@ unset PYTHONPATH unset PYTHONHOME export CUDA_PATH="$INSTALL_ENV_DIR" export CUDA_HOME="$CUDA_PATH" +export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'` # activate env bash --init-file <(echo "source \"$CONDA_ROOT_PREFIX/etc/profile.d/conda.sh\" && conda activate \"$INSTALL_ENV_DIR\"") EOF # Create start_alltalk.sh to run AllTalk cat << EOF > start_alltalk.sh #!/bin/bash +export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'` source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh" conda activate "${INSTALL_ENV_DIR}" python script.py +EOF + # Create start_finetune.sh to run AllTalk + cat << EOF > start_finetune.sh +#!/bin/bash +export TRAINER_TELEMETRY=0 +export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'` +source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh" +conda activate "${INSTALL_ENV_DIR}" +python finetune.py EOF chmod +x start_alltalk.sh chmod +x start_environment.sh + chmod +x start_finetune.sh echo echo - echo -e " start_alltalk.sh has been created." - echo -e " You can now start AllTalk with ${L_YELLOW}./start_alltalk.sh${NC}" + echo -e " Run ${L_YELLOW}./start_alltalk.sh${NC} to start AllTalk." + echo -e " Run ${L_YELLOW}./start_finetune.sh${NC} to start Finetuning." + echo -e " Run ${L_YELLOW}./start_environment.sh${NC} to start the AllTalk Python environment." + echo + echo -e " To install ${L_YELLOW}DeepSpeed${NC} on Linux, there are additional" + echo -e " steps required. Please see the Github or documentation on DeepSeed." echo read -p " Press any key to continue. " -n 1 } - delete_custom_standalone() { local env_dir="$PWD/alltalk_environment" # Check if the alltalk_environment directory exists @@ -380,47 +374,15 @@ delete_custom_standalone() { echo } -install_finetune_standalone() { - local env_dir="$PWD/alltalk_environment" - local conda_root_prefix="${env_dir}/conda" - local install_env_dir="${env_dir}/env" - if [ ! -d "${install_env_dir}" ]; then - echo " The Conda environment at '${install_env_dir}' does not exist." - echo " Please install the environment before proceeding." - read -p " Press any key to return to the menu. " -n 1 - echo - return - fi - source "${conda_root_prefix}/etc/profile.d/conda.sh" - conda activate "${install_env_dir}" - local requirements_file="requirements_finetune.txt" - echo "Installing Finetune requirements from $requirements_file..." - if ! pip install -r "$requirements_file"; then - echo " There was an error installing the Finetune requirements." - echo " Please check the output for details." - read -p " Press any key to return to the menu. " -n 1 - echo - return - fi - echo - echo " Finetune base requirements installed successfully." - echo - echo -e " Finetuning requires access to ${GREEN}Cublas${NC} from the ${GREEN}Nvidia CUDA Toolkit${NC}, so will need installing." - echo - echo -e " Instructions can be found here:" - echo -e " ${GREEN}https://github.com/erew123/alltalk_tts?#-finetuning-a-model${NC}" - echo - read -p " Press any key to continue. " -n 1 - echo -} - generate_diagnostics_standalone() { local env_dir="$PWD/alltalk_environment" local conda_root_prefix="${env_dir}/conda" local install_env_dir="${env_dir}/env" if [ ! -d "${install_env_dir}" ]; then + echo echo " The Conda environment at '${install_env_dir}' does not exist." echo " Please install the environment before proceeding." + echo read -p " Press any key to return to the menu. " -n 1 echo return @@ -428,16 +390,97 @@ generate_diagnostics_standalone() { source "${conda_root_prefix}/etc/profile.d/conda.sh" conda activate "${install_env_dir}" if ! python diagnostics.py; then + echo echo " There was an error running diagnostics." + echo read -p " Press any key to return to the menu. " -n 1 echo return fi + echo echo " Diagnostics completed successfully." read -p " Press any key to continue. " -n 1 echo } +gitpull_standalone() { + local env_dir="$PWD/alltalk_environment" + local conda_root_prefix="${env_dir}/conda" + local install_env_dir="${env_dir}/env" + if [ ! -d "${install_env_dir}" ]; then + echo + echo " The Conda environment at '${install_env_dir}' does not exist." + echo " Please install the environment before proceeding." + echo + read -p " Press any key to return to the menu. " -n 1 + echo + return + fi + source "${conda_root_prefix}/etc/profile.d/conda.sh" + conda activate "${install_env_dir}" + if ! git pull; then + echo + echo " There was an error pulling from Github." + echo + read -p " Press any key to return to the menu. " -n 1 + echo + return + fi + echo + echo " AllTalk Updated from Github. Please re-apply." + echo " the latest requirements file. (Option 3)" + echo + read -p " Press any key to continue. " -n 1 + echo +} + +pippurge_standalone() { + if ! pip cache purge; then + echo + echo " There was an error." + echo + read -p " Press any key to return to the menu. " -n 1 + echo + return + fi + echo + echo " The PIP cache has been purged." + echo + read -p " Press any key to continue. " -n 1 + echo +} + +reapply_standalone() { + local env_dir="$PWD/alltalk_environment" + local conda_root_prefix="${env_dir}/conda" + local install_env_dir="${env_dir}/env" + if [ ! -d "${install_env_dir}" ]; then + echo + echo " The Conda environment at '${install_env_dir}' does not exist." + echo " Please install the environment before proceeding." + echo + read -p " Press any key to return to the menu. " -n 1 + echo + return + fi + source "${conda_root_prefix}/etc/profile.d/conda.sh" + conda activate "${install_env_dir}" + echo + echo " Downloading and installing PyTorch. This step can take a long time" + echo " depending on your internet connection and hard drive speed. Please" + echo " be patient." + pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121 + echo + echo " Installing additional requirements." + echo + pip install -r system/requirements/requirements_standalone.txt + echo + echo " Requirements have been re-applied/updated." + echo + read -p " Press any key to continue. " -n 1 + echo +} + # Start the main menu main_menu diff --git a/diagnostics.py b/diagnostics.py index b3379ef..8bb45d1 100644 --- a/diagnostics.py +++ b/diagnostics.py @@ -8,6 +8,7 @@ try: import re import sys import glob + import site import textwrap import torch import packaging.version @@ -15,6 +16,7 @@ try: from packaging.specifiers import SpecifierSet from packaging.specifiers import SpecifierSet from packaging.version import parse as parse_version + from pathlib import Path except ImportError as e: print(f"\033[91mError importing module: {e}\033[0m\n") print("\033[94mPlease ensure you started the Text-generation-webUI Python environment with either\033[0m") @@ -22,6 +24,8 @@ except ImportError as e: print("\033[94mfrom the text-generation-webui directory, and then try running the diagnostics again.\033[0m") exit(1) +this_dir = Path(__file__).parent + try: import psutil except ImportError: @@ -30,21 +34,24 @@ except ImportError: import psutil def get_requirements_file(): - requirements_files = glob.glob("requirements*.txt") + this_dir = Path(__file__).parent # Assuming 'this_dir' is defined as the script's directory + requirements_dir = this_dir / 'system' / 'requirements' + requirements_files = list(requirements_dir.glob('requirements*.txt')) # Using pathlib for globbing + if not requirements_files: print("\033[91mNo requirements files found.\033[0m") return None print("\033[94m\nSelect a requirements file to check against (or press Enter for default 'requirements.txt'):\033[0m\n") - for i, file in enumerate(requirements_files, start=1): - print(f"{i}. {file}") + for i, file_path in enumerate(requirements_files, start=1): + print(f" {i}. {file_path.name}") # Only print the filename choice = input("\nEnter the number of your choice: ") try: choice = int(choice) - 1 - return requirements_files[choice] + return str(requirements_files[choice]) # Return the full path as a string except (ValueError, IndexError): - return "requirements.txt" + return str(requirements_dir / "requirements.txt") # Return the default path as a string # Set up logging with filemode='w' logging.basicConfig(filename='diagnostics.log', @@ -106,17 +113,23 @@ def test_cuda(): return "Fail - CUDA is not available." def find_files_in_path_with_wildcard(pattern): - # Split the system's PATH variable into a list of directories - search_path = os.environ.get('PATH', '').split(os.pathsep) - + # Get the site-packages directory of the current Python environment + site_packages_path = site.getsitepackages() found_paths = [] - # Iterate over each directory in the search path - for directory in search_path: + # Adjust the sub-directory based on the operating system + sub_directory = "nvidia/cublas" + if platform.system() == "Linux": + sub_directory = os.path.join(sub_directory, "lib") + else: + sub_directory = os.path.join(sub_directory, "bin") + # Iterate over each site-packages directory (there can be more than one) + for directory in site_packages_path: + # Construct the search directory path + search_directory = os.path.join(directory, sub_directory) # Use glob to find all files matching the pattern in this directory - for file_path in glob.glob(os.path.join(directory, pattern)): + for file_path in glob.glob(os.path.join(search_directory, pattern)): if os.path.isfile(file_path): # Ensure it's a file found_paths.append(file_path) - return found_paths # Function to log and print system information @@ -173,7 +186,7 @@ def log_system_info(): path_env = os.environ.get('PATH', 'N/A') # Check for cublas - file_name = 'cublas64_11.*' + file_name = 'cublas64_11.*' if platform.system() == "Windows" else 'libcublas.so.11*' found_paths = find_files_in_path_with_wildcard(file_name) # Compare with requirements file @@ -309,4 +322,4 @@ def log_system_info(): print(f"\033[94m\nPlease upload the log file with any support ticket.\033[0m") if __name__ == "__main__": - log_system_info() \ No newline at end of file + log_system_info() diff --git a/finetune.py b/finetune.py index 9287fd6..0b594ba 100644 --- a/finetune.py +++ b/finetune.py @@ -1,6 +1,8 @@ import argparse import os import sys +import platform +import site import tempfile import signal import gradio as gr @@ -22,7 +24,7 @@ from tqdm import tqdm from faster_whisper import WhisperModel # Use a local Tokenizer to resolve Japanese support # from TTS.tts.layers.xtts.tokenizer import multilingual_cleaners -from templates.tokenizer.tokenizer import multilingual_cleaners +from system.ft_tokenizer.tokenizer import multilingual_cleaners import importlib.metadata as metadata from packaging import version @@ -112,14 +114,21 @@ def test_cuda(): return cuda_status, cuda_icon, cuda_home def find_files_in_path_with_wildcard(pattern): - global pfc_status - # Split the system's PATH variable into a list of directories - search_path = os.environ.get('PATH', '').split(os.pathsep) + # Get the site-packages directory of the current Python environment + site_packages_path = site.getsitepackages() found_paths = [] - # Iterate over each directory in the search path - for directory in search_path: + # Adjust the sub-directory based on the operating system + sub_directory = "nvidia/cublas" + if platform.system() == "Linux": + sub_directory = os.path.join(sub_directory, "lib") + else: + sub_directory = os.path.join(sub_directory, "bin") + # Iterate over each site-packages directory (there can be more than one) + for directory in site_packages_path: + # Construct the search directory path + search_directory = os.path.join(directory, sub_directory) # Use glob to find all files matching the pattern in this directory - for file_path in glob.glob(os.path.join(directory, pattern)): + for file_path in glob.glob(os.path.join(search_directory, pattern)): if os.path.isfile(file_path): # Ensure it's a file found_paths.append(file_path) return found_paths @@ -127,7 +136,7 @@ def find_files_in_path_with_wildcard(pattern): def generate_cuda_markdown(): global pfc_status cuda_status, cuda_icon, cuda_home = test_cuda() - file_name = 'cublas64_11.*' + file_name = 'cublas64_11.*' if platform.system() == "Windows" else 'libcublas.so.11*' found_paths = find_files_in_path_with_wildcard(file_name) if found_paths: found_paths_str = ' '.join(found_paths) @@ -171,12 +180,35 @@ def get_system_ram_markdown(): ram_status_message = "Warning" if warning_if_low_ram else "" ram_status_icon = "⚠️" if warning_if_low_ram else "✅" - system_ram_markdown = f""" - ### 🟪 System RAM Information
- {ram_status_icon} **Total RAM:** {total_ram_gb:.2f} GB
- {ram_status_icon} **Available RAM:** {ram_status_message + ' - Available RAM is less than 8 GB. You have ' if warning_if_low_ram else ''} {available_ram_gb:.2f} GB available ({used_ram_percentage:.2f}% used) - """ + if torch.cuda.is_available(): + gpu_device_id = torch.cuda.current_device() + gpu_device_name = torch.cuda.get_device_name(gpu_device_id) + # Get the total and available memory in bytes, then convert to GB + gpu_total_mem_gb = torch.cuda.get_device_properties(gpu_device_id).total_memory / (1024 ** 3) + # gpu_available_mem_gb = (torch.cuda.get_device_properties(gpu_device_id).total_memory - torch.cuda.memory_allocated(gpu_device_id)) / (1024 ** 3) + # gpu_available_mem_gb = (torch.cuda.get_device_properties(gpu_device_id).total_memory - torch.cuda.memory_reserved(gpu_device_id)) / (1024 ** 3) + gpu_reserved_mem_gb = torch.cuda.memory_reserved(gpu_device_id) / (1024 ** 3) + gpu_available_mem_gb = gpu_total_mem_gb - gpu_reserved_mem_gb + # Check if total or available memory is less than 11 GB and set icons + gpu_total_status_icon = "⚠️" if gpu_total_mem_gb < 12 else "✅" + gpu_available_status_icon = "⚠️" if gpu_available_mem_gb < 12 else "✅" + gpu_status_icon = "✅" + else: + gpu_status_icon = "⚠️" + gpu_device_name = "Cannot detect a CUDA card" + gpu_total_mem_gb = "Cannot detect a CUDA card" + gpu_available_mem_gb = "Cannot detect a CUDA card" + gpu_total_status_icon = gpu_status_icon + gpu_available_status_icon = gpu_status_icon + system_ram_markdown = f""" + ### 🟪 System RAM and VRAM Information
+ {ram_status_icon} **Total RAM:** {total_ram_gb:.2f} GB
+ {ram_status_icon} **Available RAM:** {ram_status_message + ' - Available RAM is less than 8 GB. You have ' if warning_if_low_ram else ''} {available_ram_gb:.2f} GB available ({used_ram_percentage:.2f}% used)
+ {gpu_status_icon} **GPU Name:** {gpu_device_name}
+ {gpu_total_status_icon} **GPU Total RAM:** {gpu_total_mem_gb:.2f} GB
+ {gpu_available_status_icon} **GPU Available RAM:** {gpu_available_mem_gb:.2f} GB
+ """ return system_ram_markdown @@ -295,6 +327,7 @@ def format_audio_list(target_language, whisper_model, out_path, eval_split_numbe device = "cuda" if torch.cuda.is_available() else "cpu" print("[FINETUNE] Loading Whisper Model:", whisper_model) + print("[FINETUNE] Model will be downloaded if its not available, which will take a few minutes.") asr_model = WhisperModel(whisper_model, device=device, compute_type="float32") metadata = {"audio_file": [], "text": [], "speaker_name": []} @@ -472,6 +505,8 @@ def format_audio_list(target_language, whisper_model, out_path, eval_split_numbe # deallocate VRAM and RAM del asr_model, final_eval_set, final_training_set, new_data_df, existing_metadata gc.collect() + if torch.cuda.is_available(): + torch.cuda.empty_cache() existing_train_df = None existing_eval_df = None print("[FINETUNE] Train CSV:", train_metadata_path) @@ -512,6 +547,24 @@ def train_gpt(language, num_epochs, batch_size, grad_acumm, train_csv, eval_csv, print(f"[FINETUNE] \033[94mLanguage: \033[92m{language} \033[94mEpochs: \033[92m{num_epochs} \033[94mBatch size: \033[92m{batch_size}\033[0m \033[94mGrad accumulation steps: \033[92m{grad_acumm}\033[0m") print(f"[FINETUNE] \033[94mTraining : \033[92m{train_csv}\033[0m") print(f"[FINETUNE] \033[94mEvaluation : \033[92m{eval_csv}\033[0m") + if torch.cuda.is_available(): + torch.cuda.empty_cache() + # Get the current device ID + gpu_device_id = torch.cuda.current_device() + gpu_available_mem_gb = (torch.cuda.get_device_properties(gpu_device_id).total_memory - torch.cuda.memory_allocated(gpu_device_id)) / (1024 ** 3) + print(f"[FINETUNE] \033[94mAvailable VRAM: \033[92m{gpu_available_mem_gb:.2f} GB\033[0m") + if gpu_available_mem_gb < 12: + print(f"[FINETUNE]") + print(f"[FINETUNE] \033[91m****** WARNING PRE-FLIGHT CHECKS FAILED ******* WARNING PRE-FLIGHT CHECKS FAILED *****\033[0m") + print(f"[FINETUNE] \033[94mAvailable VRAM: \033[92m{gpu_available_mem_gb:.2f} GB\033[0m") + print(f"[FINETUNE] \033[94mIf you are running on a Linux system and you have 12GB's or less of VRAM, this step\033[0m") + print(f"[FINETUNE] \033[94mmay fail, due to not enough GPU VRAM. Windows systems will use system RAM as extended\033[0m") + print(f"[FINETUNE] \033[94mVRAM and so should work ok. However, Windows machines will need enough System RAM\033[0m") + print(f"[FINETUNE] \033[94mavailable. Please read the PFC help section available on the first tab of the web\033[0m") + print(f"[FINETUNE] \033[94minterface for more information.\033[0m") + print(f"[FINETUNE] \033[91m****** WARNING PRE-FLIGHT CHECKS FAILED ******* WARNING PRE-FLIGHT CHECKS FAILED *****\033[0m") + print(f"[FINETUNE]") + # Create the directory os.makedirs(OUT_PATH, exist_ok=True) @@ -1111,14 +1164,17 @@ if __name__ == "__main__": with gr.Tab("🟪 RAM & VRAM Help"): gr.Markdown( f""" - {system_ram_results}
- ◽ During actual training (Step 2) Finetuning will use around 14GB of VRAM. If your GPU doesnt have 14GB's of VRAM it will attempt to overspill into your System RAM. So if you also have very low or slow System RAM you can expect bad performance or the training to fail.
- ◽ 12GB cards may need 2GB System RAM.
- ◽ 8GB cards may need 6GB System RAM.
- ◽ 6GB cards may need 8GB System RAM.
+ {system_ram_results}
+ ◽ During actual training (Step 2) Finetuning will use around **14GB's** of VRAM. If your GPU doesnt have 14GB's of VRAM:
+ - **Windows** systems will attempt to extend VRAM into your System RAM, and so should work.
+ - **Linux** systems can only use the available VRAM, so may fail on 12GB VRAM or smaller GPU's.
+ ◽ For **Windows** users with 12GB or less, if you also have very low or slow System RAM you can expect bad performance or the training to fail.
+ - 12GB cards may need 2GB System RAM.
+ - 8GB cards may need 6GB System RAM.
+ - 6GB cards may need 8GB System RAM.
◽ Its hard to estimate what performance impact this could will have, due to different memory speeds, PCI speeds, GPU speeds etc.
◽ If you have a low VRAM scenario and also are attempting to run on a mechanical hard drive, Finetuning could take ??? amount of time.
- ◽ On Windows machines, please ensure you have not disabled System Memory Fallback for Stable Diffusion link here
+ ◽ On Windows machines, please ensure you have **not** disabled System Memory Fallback for Stable Diffusion link here
""" ) with gr.Tab("🟨 CUDA & Cublas Help"): @@ -1437,6 +1493,7 @@ if __name__ == "__main__": ◽ Audio clips shorter than 7 seconds are automatically excluded from the reference list, as they generally do not provide sufficient length for effective TTS generation.
◽ Should you find a lower number of reference files than anticipated, it may be due to the initial segmentation performed by Whisper (Step 1), which can occasionally result in clips that are too brief for our criteria or indeed too long. In such cases, consider manually segmenting your audio to give Whisper a better chance at generating training data and repeating the process.
### 🟦 What you need to do + ◽ **Click** the **Refresh Dropdowns** button to correctly populate the **Speaker Reference Audio** with all the available WAV samples.
◽ The model is now trained and you are at the testing stage. Hopefully all the dropdowns should be pre-populated now.
◽ You need to Load Fine-tuned XTTS model and then select your Speaker Reference Audio. You can choose various Speaker Reference Audios to see which works best.
◽ All the Speaker Reference Audios in the dropdown are ones that are 8 seconds long or more. You can use one of these later for your voice sample in All Talk, so remember the one you like.
diff --git a/modeldownload.py b/modeldownload.py index 971ae4a..b4edae5 100644 --- a/modeldownload.py +++ b/modeldownload.py @@ -45,18 +45,18 @@ else: ) # Read the version specifier from requirements_nvidia.txt -with open(this_dir / "requirements_nvidia.txt", "r") as req_file: +with open(this_dir / "system" / "requirements" / "requirements_standalone.txt", "r") as req_file: requirements = req_file.readlines() tts_version_required = None for req in requirements: - if req.startswith("TTS>="): - tts_version_required = req.strip().split(">=")[1] + if req.startswith("TTS=="): + tts_version_required = req.strip().split("==")[1] break if tts_version_required is None: raise ValueError( - f"[{params['branding']}Startup] \033[91mWarning\033[0m Could not find TTS version specifier in requirements_nvidia.txt" + f"[{params['branding']}Startup] \033[91mWarning\033[0m Could not find TTS version specifier in requirements file" ) diff --git a/script.py b/script.py index 3c442af..5b4066d 100644 --- a/script.py +++ b/script.py @@ -85,7 +85,7 @@ except ModuleNotFoundError: #### STARTUP VARIABLES #### ########################### # STARTUP VARIABLE - Import languges file for Gradio to be able to display them in the interface -with open(this_dir / "languages.json", encoding="utf8") as f: +with open(this_dir / "system" / "config" / "languages.json", encoding="utf8") as f: languages = json.load(f) # Create a global lock process_lock = threading.Lock() @@ -347,26 +347,26 @@ def send_reload_request(tts_method): params["tts_method_xtts_local"] = False params["tts_method_api_tts"] = True params["deepspeed_activate"] = False - audio_path = this_dir / "templates" / "apitts.wav" + audio_path = this_dir / "system" / "at_sounds" / "apitts.wav" tts_method_xtts_ft = False elif tts_method == "API Local": params["tts_method_api_tts"] = False params["tts_method_xtts_local"] = False params["tts_method_api_local"] = True params["deepspeed_activate"] = False - audio_path = this_dir / "templates" / "apilocal.wav" + audio_path = this_dir / "system" / "at_sounds" / "apilocal.wav" tts_method_xtts_ft = False elif tts_method == "XTTSv2 Local": params["tts_method_api_tts"] = False params["tts_method_api_local"] = False params["tts_method_xtts_local"] = True - audio_path = this_dir / "templates" / "xttslocal.wav" + audio_path = this_dir / "system" / "at_sounds" / "xttslocal.wav" tts_method_xtts_ft = False elif tts_method == "XTTSv2 FT": params["tts_method_api_tts"] = False params["tts_method_api_local"] = False params["tts_method_xtts_local"] = False - audio_path = this_dir / "templates" / "xttsfinetuned.wav" + audio_path = this_dir / "system" / "at_sounds" / "xttsfinetuned.wav" tts_method_xtts_ft = True return f'' except requests.exceptions.RequestException as e: @@ -385,9 +385,9 @@ def send_lowvram_request(low_vram): try: params["tts_model_loaded"] = False if low_vram: - audio_path = this_dir / "templates" / "lowvramenabled.wav" + audio_path = this_dir / "system" / "at_sounds" / "lowvramenabled.wav" else: - audio_path = this_dir / "templates" / "lowvramdisabled.wav" + audio_path = this_dir / "system" / "at_sounds" / "lowvramdisabled.wav" url = f"{base_url}/api/lowvramsetting?new_low_vram_value={low_vram}" headers = {"Content-Type": "application/json"} response = requests.post(url, headers=headers) @@ -414,9 +414,9 @@ def send_deepspeed_request(deepspeed_param): try: params["tts_model_loaded"] = False if deepspeed_param: - audio_path = this_dir / "templates" / "deepspeedenabled.wav" + audio_path = this_dir / "system" / "at_sounds" / "deepspeedenabled.wav" else: - audio_path = this_dir / "templates" / "deepspeeddisabled.wav" + audio_path = this_dir / "system" / "at_sounds" / "deepspeeddisabled.wav" url = f"{base_url}/api/deepspeed?new_deepspeed_value={deepspeed_param}" headers = {"Content-Type": "application/json"} @@ -511,7 +511,7 @@ def combine(audio_files, output_folder, state): ################################ # PREVIEW VOICE - Generate Random Sentence if Voice Preview box is empty def random_sentence(): - with open(this_dir / "harvard_sentences.txt") as f: + with open(this_dir / "system" / "config" / "harvard_sentences.txt") as f: return random.choice(list(f)) diff --git a/system/admin.html b/system/admin.html new file mode 100644 index 0000000..90a85e6 --- /dev/null +++ b/system/admin.html @@ -0,0 +1,2614 @@ + + + + + + + +
AllTalk TTS
+ ++
Page Index
+-
+
- 🔊 Using Voice Samples +
- 🧮 Low VRAM +
- 🏰 SillyTavern Support +
- 🟩 Good to know +
- 🟪 Updating & problems with updating +
- 🔵🟢 DeepSpeed + Installation (Windows & Linux) +
- 🆘 Support + Requests, Troubleshooting & Feature requests +
- 🟨 Help with problems +
- ⚫ Finetuning a model +
- ⬜ AllTalk TTS Generator +
- 🟠 API Suite and + JSON-CURL + +
Links
+ +-
+
- AllTalk TTS Generator +
- AllTalk Github + +
- AllTalk + Changelog +
- AllTalk + Existing Feature Requests +
- AllTalk + Help +
AllTalk Server Status
+-
+
- Base URL:
http://{{ params.ip_address }}:{{ params.port_number }}
+ - Server Status:
+
http://{{ params.ip_address }}:{{ params.port_number }}/ready+
+
+ +
🛠️ About this project
+AllTalk is a labour of love, developed and supported in my personal free time. As such, my ability to respond to + support requests is limited. I prioritize issues based on their impact and the number of users affected. I + appreciate your understanding and patience. If your inquiry isn't covered by the documentation or existing + discussions, and it's not related to a bug or feature request, I'll do my best to assist as time allows. +
++
💖 Showing Your Support
+If AllTalk has been helpful to you, consider showing your support through a donation on my Ko-fi page. Your support is greatly appreciated and helps ensure the + continued development and improvement of AllTalk.
++ + +
🗣️ Demo/Test TTS
++ +
🔊 Using Voice Samples
+Where are the sample voices stored?
+Voice samples are stored in /alltalk_tts/voices/ + and should be named using the following format name.wav
+ +🔊 Where are the outputs stored & Automatic output wav file deletion.
+Voice outputs are stored in /alltalk_tts/outputs/
+You can configure automatic maintenance deletion of old wav + files by setting Del WAV's older than in the settings above.
+When Disabled your output + wav files will be left untouched. When set to a setting 1 Day or greater, + your output wav files older than that time period will be automatically deleted on start-up of AllTalk.
+ +🔊 Where are the models stored?
+This extension will download the 2.0.2 model to /alltalk_tts/models/
+This TTS engine will also download the latest available model + and store it wherever your OS normally stores it (Windows/Linux/Mac).
+ +🔊 How do I create a new voice sample?
+To create a new voice sample, you need to make a wav file that + is + 22050Hz, Mono, 16 bit and between 6 to 30 seconds long, though 8 to 10 seconds is usually + good enough. The model can handle up to 30 second samples, however I've not noticed any improvement in voice + output from a much longer clips. +
+You want to find a nice clear selection of audio, so lets say + you wanted to clone your favourite celebrity. You may go looking for an interview where they are talking. Pay + close attention to the audio you are listening to and trying to sample. Are there noises in the background, hiss + on the soundtrack, a low humm, some quiet music playing or something? The better quality the audio the better + the final TTS result. Don't forget, the AI that processes the sounds can hear everything in your sample and it + will use them in the voice its trying to recreate.
+Try make your clip one of nice flowing speech, like the included + example files. No big pauses, gaps or other sounds. Preferably a sample that the person you are trying to copy + will show a little vocal range and emotion in their voice. Also, try to avoid a clip starting or ending with + breathy sounds (breathing in/out etc).
+ +🔊 Editing your sample!
+So, you’ve downloaded your favourite celebrity interview off + YouTube, from here you need to chop it down to 6 to 30 seconds in length and resample it.
+If you need to clean it up, do audio processing, volume level
+ changes etc, do this before down-sampling.
Using the latest version of Audacity select/highlight your 6 to 30 second clip and:
Tracks
+ > Resample to 22050Hz then
Tracks >
+ Mix > Stereo to
+ Mono then
+ File > Export Audio saving it as a WAV of 22050Hz.
+
Save your generated wav file in the /alltalk_tts/voices/ folder.
+Its worth mentioning that using AI generated audio clips and bad quality samples may + introduce unwanted sounds into TTS generation.
+ +🔊 Why doesnt it sound like XXX Person?
+Maybe you might be interested in trying Finetuning of the model. Otherwise, + the reasons + can be that you:
+-
+
- Didn't down-sample it as above. +
- Have a bad quality voice sample. +
- Try using the 3x different generation methods: API TTS, API Local, and XTTSv2 Local within + the web interface, as they generate output in different ways and sound different. +
Some samples just never seem to work correctly, so maybe try a + different sample. Always remember though, this is an AI model attempting to re-create a voice, so you will never + get a 100% match.
+ + ++
🧮 Low VRAM
+The Low VRAM option is a crucial feature designed to enhance + performance under constrained (VRAM) conditions, as the TTS models require 2GB-3GB of VRAM to run effectively. + This feature strategically manages the relocation of the Text-to-Speech (TTS) model between your system's Random + Access Memory (RAM) and VRAM, moving it between the two on the fly. Obviously, this is very useful for people + who have smaller graphics cards or people whos LLM has filled their VRAM.
+When you don't have enough VRAM free after loading your LLM + model + into your VRAM (Normal Mode example below), you can see that with so little working space, your GPU will have to + swap in and out bits of the TTS model, which causes horrible slowdown.
+Note: An Nvidia Graphics + card is required for the LowVRAM option to work, as you will just be using system RAM otherwise.
+ +🧮 How It Works
+The Low VRAM mode intelligently orchestrates the relocation of + the entire TTS model and stores the TTS model in your system RAM. When the TTS engine requires VRAM for + processing, the entire model seamlessly moves into VRAM, causing your LLM to unload/displace some layers, + ensuring optimal performance of the TTS engine.
+Post-TTS processing, the model moves back to system RAM, freeing + up VRAM space for your Language Model (LLM) to load back in the missing layers. This adds about 1-2 seconds to + both text generation by the LLM and the TTS engine.
+By transferring the entire model between RAM and VRAM, the Low + VRAM option avoids fragmentation, ensuring the TTS model remains cohesive and has all the working space it needs + in your GPU, without having to just work on small bits of the TTS model at a time (which causes terrible slow + down).
+This creates a TTS generation performance Boost for Low VRAM + Users and is particularly beneficial for users with less than 2GB of free VRAM after loading their LLM, + delivering a substantial 5-10x improvement in TTS generation speed.
+🧮 Visual explanation
+
+ + +
🏰 SillyTavern Support
+Important note for Text-generation-webui users
+You HAVE to disable + Enable TTS within the Text-generation-webui AllTalk interface, + otherwise Text-generation-webui will also generate TTS due to the way it sends out text. You can do this + each time you start up Text-generation-webui or set it in the start-up settings at the top of this page. +
+🏰 Quick Tips
+-
+
- Only change DeepSpeed, Low VRAM or Model one at a time. Wait for it to say Ready before changing something + else. +
- You can permanently change the AllTalk startup settings or DeepSpeed, Low VRAM and Model at the top of this + page. +
- Different AI models use quotes and asterisks differently, so you may need to change "Text not inside" + depending on model. +
- Add new voice samples to the voices folder. You can Finetune a model to make it sound even closer to the + original sample. +
- DeepSpeed will improve processing time to TTS to be 2-3x faster. +
- Low VRAM can be very beneficial if you don't have much memory left after loading your LLM. +
🏰 TTS Generation Methods in SilllyTavern
+You have 2 types of audio generation options, Streaming and Standard.
+The Streaming Audio Generation method is designed for speed and + is best suited for situations where you just want quick audio playback. This method, however, is limited to + using just one voice per TTS generation request. This means a limitation of the Streaming method is the + inability to utilize the AllTalk narrator function, making it a straightforward but less nuanced option.
+On the other hand, the Standard Audio Generation method provides + a richer auditory experience. It's slightly slower than the Streaming method but compensates for this with its + ability to split text into multiple voices. This functionality is particularly useful in scenarios where + differentiating between character dialogues and narration can enhance the storytelling and delivery. The + inclusion of the AllTalk narrator functionality in the Standard method allows for a more layered and immersive + experience, making it ideal for content where depth and variety in voice narration add significant value.
+In summary, the choice between Streaming and Standard methods in + AllTalk TTS depends on what you want. Streaming is great for quick and simple audio generation, while Standard + is preferable for a more dynamic and engaging audio experience.
+-
+
- AllTalk TTS Generation Method:
+
-
+
- Select between Standard and Streaming Audio Generation methods. +
- This setting impacts the AllTalk narrator functionality. +
+ - Language Selection:
+-
+
- Select your preferred TTS generation language from the "Language" dropdown. +
+ - Model Switching:
+
-
+
- Switch between different TTS models like API TTS, API Local, XTTSv2 Local, and optionally XTTSv2 FT + if you have a finetuned model available. +
- Fine-tuned model availability (XTTSv2 FT) will only show when a finetuned model is detected by + AllTalk. +
- See TTS Models/Methods for more information (though most people will want to stick with XTTSv2 + Local). +
+
+
+ - DeepSpeed and Low VRAM Options:
+
-
+
- Optimize performance with DeepSpeed and Low VRAM settings. +
- DeepSpeed can offer a 2-3x performance boost on TTS generation. (Requires installation) +
- See the relevant sections in this documentation for details. +
+
Changing model or DeepSpeed or Low VRAM each take about 15 + seconds so you should only change one at + a time and wait for Ready before changing the next setting. To set these + options long term you can apply the settings at the top of this page.
+🏰 AllTalk Narrator
+Only available on the Standard Audo Generation method.
+-
+
- Narrator Voice Selection:
+
-
+
- Allows users to choose different narrator voices. +
- Access via the "Narrator Voice" dropdown. +
+ - AllTalk Narrator:
+
-
+
- Toggle the AllTalk narrator feature. +
- Access via "AT Narrator" dropdown with Enabled/Disabled options. +
+ - Text Outside Asterisks Handling:
+
-
+
- Choose how text outside asterisks is interpreted (as Narrator or Character voice). +
- Managed via the "Text Not Inside * or "" dropdown. +
- Note, only available when the AllTalk Narrator is enabled. +
+
🏰 Usage Notes:
+-
+
- On startup of SillyTavern, it will pull your current settings from AllTalk (Current model, DeepSpeed status, + Low VRAM status and Finetuned model availability). +
- Enabling the narrator automatically unchecks certain checkboxes related to text handling. +
- Changes in model or settings might trigger multiple requests to the server; patience is advised. +
🏰 Troubleshooting:
+-
+
- If experiencing issues, use the Reload button in SillyTavern's TTS extension to reinitialize the connection + to AllTalk and check if AllTalk is started correctly. +
+
🟩 Good to know
+ +🟩 A note on Character Cards, Greeting Messages & + Text-not-inside
+Click to expand
+ +Messages intended for the Narrator should be enclosed in asterisks * and those for the character
+ inside quotation marks ". However, AI systems often deviate from these rules, resulting in
+ text that is neither in quotes nor asterisks. Sometimes, text may appear with only a single asterisk, and AI
+ models may vary their formatting mid-conversation. For example, they might use asterisks initially and then
+ switch to unmarked text. A properly formatted line should look like this:
"Hey! I'm so excited to finally meet you. I've heard so many great things about you
+ and I'm eager to pick your brain about computers." *She walked across the
+ room and picked up her cup of coffee*
Most narrator/character systems switch voices upon encountering an asterisk or quotation marks, which is + somewhat effective. AllTalk has undergone several revisions in its sentence splitting and identification + methods. While some irregularities and AI deviations in message formatting are inevitable, any line + beginning or ending with an asterisk should now be recognized as Narrator dialogue. Lines enclosed in double + quotes are identified as Character dialogue. For any other text, you can choose how AllTalk handles it: + whether it should be interpreted as Character or Narrator dialogue (most AI systems tend to lean more + towards one format when generating text not enclosed in quotes or asterisks).
+With improvements to the splitter/processor, I'm confident it's functioning well. You can monitor + what AllTalk identifies as Narrator lines on the command line and adjust its behavior if needed (Text Not + Inside - Function).
+
🟩 Changing AllTalks IP address & + Accessing AllTalk over your Network
+Click to expand
+ +AllTalk is coded to start on 127.0.0.1, meaning that it will ONLY be accessable to the local computer it is + running on. If you want to make AllTalk available to other systems on your network, you will need to change + its IP address to match the IP address of your network card/computers current IP address. There are 2x ways + to change the IP address:
+-
+
- Start AllTalk and within its web interface and you can edit the IP address on the "AllTalk Startup + Settings". +
- You can edit the
confignew.jsonfile in a text editor and change +"ip_address": "127.0.0.1",to the IP address of your choosing. +
+
So, for example, if your computer's network card was on IP address 192.168.0.20, you would change + AllTalk's setting to 192.168.1.20 and then restart AllTalk. You will need to ensure + your machine stays on this IP address each time it is restarted, by setting your machine to have a static IP + address.
+🟩 Text-geneneration-webui + & Stable-Diffusion Plugin - Load Order & stripped text
+Click to expand
+ +The Stable Diffusion plugin for Text-generation-webui strips out some of the text, which is
+ passed to Stable Diffusion for image/scene generation. Because this text is stripped, its important to
+ consider the load order of the plugins to get the desired result you want. Lets assume the AI has just
+ generated the following message
+ *He walks into the room with a smile on his face and says* Hello how are you?. Depending on the
+ load order will change what text reaches AllTalk for generation e.g.
+
SD Plugin loaded before AllTalk - Only Hi how are you? is sent to AllTalk, with
+ the *He walks into the room with a smile on his face and says* being sent over to SD for image
+ generation. Narration of the scene is not possible.
+ AllTalk loaded before SD Plugin -
+ *He walks into the room with a smile on his face and says* Hello how are you? is sent to
+ AllTalk with the *He walks into the room with a smile on his face and says* being sent over to
+ SD for image generation.
+ The load order can be changed within Text-generation-webui's settings.yaml file or
+ cmd_flags.txt (depending on how you are managing your extensions).
+ 
+
🟩 I want to know more about the XTTS AI model used +
+Click to expand
+ +Currently the XTTS model is the main model used by AllTalk for TTS generation. If you want to know more + details about the XTTS model, its capabilties or its technical features you can look at resources such as: +
+ ++
🟪 Updating
+Maintaining the latest version of your setup ensures access to new features and improvements. Below are the steps + to update your installation, whether you're using Text-Generation-webui or running as a Standalone + Application.
+UPDATING - Text-Generation-webui
+ +The update process closely mirrors the installation steps. Follow these to ensure your setup remains current: +
+-
+
-
+
Open a Command Prompt/Terminal:
+-
+
- Navigate to your Text-Generation-webui folder with:
-
+
cd text-generation-webui
+
+
+ - Navigate to your Text-Generation-webui folder with:
-
+
Start the Python Environment:
+-
+
- Activate the Python environment tailored for your operating system. Use the appropriate command
+ from below based on your OS:
-
+
- Windows:
cmd_windows.bat
+ - Linux:
./cmd_linux.sh
+ - macOS:
cmd_macos.sh
+ - WSL (Windows Subsystem for Linux):
cmd_wsl.bat
+
+ - Windows:
+
+If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
+ - Activate the Python environment tailored for your operating system. Use the appropriate command
+ from below based on your OS:
-
+
Navigate to the AllTalk TTS Folder:
+-
+
- Move into your extensions and then the alltalk_tts directory:
-
+
cd extensions/alltalk_tts
+
+
+ - Move into your extensions and then the alltalk_tts directory:
-
+
Update the Repository:
+-
+
- Fetch the latest updates from the repository with:
-
+
git pull
+
+
+ - Fetch the latest updates from the repository with:
-
+
Install Updated Requirements:
+-
+
- Depending on your machine's OS, install the required dependencies using pip:
-
+
- For Windows Machines:
-
+
pip install -r system\requirements\requirements_textgen.txt
+
+ - For Linux/Mac:
-
+
pip install -r system/requirements/requirements_textgen.txt
+
+
+ - For Windows Machines:
+ - Depending on your machine's OS, install the required dependencies using pip:
UPDATING - Standalone Application
+ +For Standalone Application users, here's how to update your setup:
+-
+
-
+
Open a Command Prompt/Terminal:
+-
+
- Navigate to your AllTalk folder with:
-
+
cd alltalk_tts
+
+
+ - Navigate to your AllTalk folder with:
-
+
Access the Python Environment:
+-
+
- In a command prompt or terminal window, navigate to your
alltalk_ttsdirectory and + start the Python environment:-
+
- Windows:
-
+
start_environment.bat
+
+ - Linux/macOS:
-
+
./start_environment.sh
+
+
+ - Windows:
+ - In a command prompt or terminal window, navigate to your
++If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
-
+
-
+
Pull the Latest Updates:
+-
+
- Retrieve the latest changes from the repository with:
-
+
git pull
+
+
+ - Retrieve the latest changes from the repository with:
-
+
Install Updated Requirements:
+-
+
- Depending on your machine's OS, install the required dependencies using pip:
-
+
- For Windows Machines:
-
+
pip install -r system\requirements\requirements_standalone.txt
+
+ - For Linux/Mac:
-
+
pip install -r system/requirements/requirements_standalone.txt
+
+
+ - For Windows Machines:
+ - Depending on your machine's OS, install the required dependencies using pip:
🟪 Resolving Update Issues
+If you encounter problems during or after an update, following these steps can help resolve the issue by + refreshing your installation while preserving your data:
+RESOLVING - Text-Generation-webui
+ +The process involves renaming your existing alltalk_tts directory, setting up a fresh instance,
+ and then migrating your data:
-
+
-
+
Rename Existing Directory:
+-
+
- First, rename your current
alltalk_ttsfolder to back it up, e.g., +alltalk_tts.old. This preserves any existing data. +
+
+ - First, rename your current
-
+
Open a Console/Terminal:
+-
+
-
+
Navigate to the Text-generation-webui directory and start the Python environment appropriate + for your operating system:
+-
+
cd text-generation-webui
+
+ -
+
Then use one of the following commands based on your OS:
+-
+
- Windows:
cmd_windows.bat
+ - Linux:
./cmd_linux.sh
+ - macOS:
cmd_macos.sh
+ - WSL (Windows Subsystem for Linux):
cmd_wsl.bat
+
+
+If you're not familiar with Python environments, see Understanding Python + Environments Simplified in the Help section for more info.
+
+ - Windows:
+ -
+
-
+
Clone the AllTalk TTS Repository:
+-
+
- Move into the
extensionsdirectory and clone a fresh copy of +alltalk_tts:-
+
cd extensions
+ git clone https://github.com/erew123/alltalk_tts
+
+
+ - Move into the
-
+
Install Requirements:
+-
+
- Navigate to the newly cloned
alltalk_ttsdirectory and install the necessary + dependencies for your system:-
+
cd alltalk_tts
+
+ - Depending on your machine's OS, install the required dependencies using pip:
-
+
- For Windows Machines:
-
+
pip install -r system\requirements\requirements_textgen.txt
+
+ - For Linux/Mac:
-
+
pip install -r system/requirements/requirements_textgen.txt
+
+
+ - For Windows Machines:
+ - Navigate to the newly cloned
-
+
Migrate Your Data:
+-
+
- Before starting the application, transfer the
models,voices, and +outputsfolders fromalltalk_tts.oldto the new +alltalk_ttsdirectory. This action preserves your voice history and prevents the + need to re-download the model. +
+
+ - Before starting the application, transfer the
You're now ready to launch Text-generation-webui. Note that you may need to reapply any previously saved + configuration changes through the configuration page.
+-
+
- Final Step: +
-
+
- Once you've verified that everything is working as expected and you're satisfied with the setup,
+ feel free to delete the
alltalk_tts.olddirectory to free up space.
+
+
🔵🟢 DeepSpeed Installation Options
+DeepSpeed requires an Nvidia Graphics card
+DeepSpeed provides a 2x-3x speed boost for Text-to-Speech and AI + tasks. It's all about making AI and TTS happen faster and more efficiently.
+🔵 Linux Installation
+DeepSpeed requires access to the Nvidia CUDA Development Toolkit to compile on a Linux system. + It's important to note that this toolkit is distinct and unrealted to your graphics card driver or the CUDA + version the Python environment uses.
+Linux DeepSpeed - Text-generation-webui
+ +DeepSpeed Installation for Text generation webUI
+-
+
-
+
Nvidia CUDA Development Toolkit Installation:
+-
+
- The toolkit is crucial for DeepSpeed to compile/build for your version of Linux and requires + around 3GB's of disk space. +
- Install using your package manager (Recommended) e.g. CUDA Toolkit + 11.8 or download directly from Nvidia CUDA Toolkit Archive + (choose 11.8 or 12.1 for Linux). +
+ -
+
Open a Terminal Console:
+-
+
- After Nvidia CUDA Development Toolkit installation, access your terminal console. +
+ -
+
Install libaio-dev:
+-
+
-
+
Use your Linux distribution's package manager.
+
+ -
+
-
+
sudo apt install libaio-devfor Debian-based systems
+ sudo yum install libaio-develfor RPM-based systems.
+
+
+ -
+
-
+
Navigate to Text generation webUI Folder:
+-
+
- Change directory to your Text generation webUI folder with
+
cd text-generation-webui. +
+
+ - Change directory to your Text generation webUI folder with
+
-
+
Activate Text generation webUI Custom Conda Environment:
+-
+
- Run
./cmd_linux.shto start the environment.
+
+
+If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
+ - Run
-
+
Set
+CUDA_HOMEEnvironment Variable:-
+
- DeepSpeed locates the Nvidia toolkit using the
CUDA_HOMEenvironment variable.
+ - You will only set this temporarily as Text generation webUI sets up its own CUDA_HOME
+ environment each time you use
./cmd_linux.shor./start_linux.sh
+
+ - DeepSpeed locates the Nvidia toolkit using the
-
+
Temporarily Configuring
+CUDA_HOME:-
+
-
+
When the Text generation webUI Python environment is active (step 5), set +
+CUDA_HOME. +
+ -
+
-
+
export CUDA_HOME=/usr/local/cuda
+ export PATH=${CUDA_HOME}/bin:${PATH}
+ export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+
+ -
+
You can confirm the path is set correctly and working by running the command +
+nvcc --versionshould confirm +Cuda compilation tools, release 11.8.. +
+ -
+
Incorrect path settings may lead to errors. If you encounter path issues or receive errors + like
+[Errno 2] No such file or directorywhen you run the next step, confirm + the path correctness or adjust as necessary.
+
+ -
+
-
+
DeepSpeed Installation:
+-
+
- Install DeepSpeed using
pip install deepspeed.
+
+ - Install DeepSpeed using
-
+
Troubleshooting:
+-
+
- Troubleshooting steps for DeepSpeed installation can be located down below. +
- NOTE: You DO NOT need to set Text-generation-webUI's + --deepspeed setting for AllTalk to be able to use DeepSpeed. These are two + completely separate things and incorrectly setting that on Text-generation-webUI may cause other + complications. +
Linux DeepSpeed - Standalone Installation
+ + + + +DeepSpeed Installation for Standalone AllTalk
+-
+
-
+
Nvidia CUDA Development Toolkit Installation:
+-
+
- The toolkit is crucial for DeepSpeed to compile/build for your version of Linux and requires + around 3GB's of disk space. +
- Install using your package manager (Recommended) e.g. CUDA Toolkit + 11.8 or download directly from Nvidia CUDA Toolkit Archive + (choose 11.8 or 12.1 for Linux). +
+ -
+
Open a Terminal Console:
+-
+
- After Nvidia CUDA Development Toolkit installation, access your terminal console. +
+ -
+
Install libaio-dev:
+-
+
-
+
Use your Linux distribution's package manager.
+
+ -
+
-
+
sudo apt install libaio-devfor Debian-based systems
+ sudo yum install libaio-develfor RPM-based systems.
+
+
+ -
+
-
+
Navigate to AllTalk TTS Folder:
+-
+
- Change directory to your AllTalk TTS folder with
cd alltalk_tts.
+
+ - Change directory to your AllTalk TTS folder with
-
+
Activate AllTalk Custom Conda Environment:
+-
+
- Run
./start_environment.shto start the AllTalk Python environment.
+ - This command will start the custom Python environment that was installed with
+
./atsetup.sh.
+
+
+
+If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
+ - Run
-
+
Set
+CUDA_HOMEEnvironment Variable:-
+
- The DeepSpeed installation routine locates the Nvidia toolkit using the
CUDA_HOME+ environment variable. This can be set temporarily for a session or permanently, depending on + other requirements you may have for other Python/System environments.
+ - For temporary use, proceed to step 8. For a permanent solution, see Conda's + manual on setting environment variables. +
+ - The DeepSpeed installation routine locates the Nvidia toolkit using the
-
+
(Optional) Permanent
+CUDA_HOMESetup:-
+
- If you choose to set
CUDA_HOMEpermanently, follow the instructions in the provided + Conda manual link above.
+
+ - If you choose to set
-
+
Configuring
+CUDA_HOME:-
+
-
+
When your Python environment is active (step 5), set
+CUDA_HOME. +
+ -
+
-
+
export CUDA_HOME=/usr/local/cuda
+ export PATH=${CUDA_HOME}/bin:${PATH}
+ export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+
+ -
+
You can confirm the path is set correctly and working by running the command +
+nvcc --versionshould confirm +Cuda compilation tools, release 11.8.. +
+ -
+
Incorrect path settings may lead to errors. If you encounter path issues or receive errors + like
+[Errno 2] No such file or directorywhen you run the next step, confirm + the path correctness or adjust as necessary.
+
+ -
+
-
+
DeepSpeed Installation:
+-
+
- Install DeepSpeed using
pip install deepspeed.
+
+ - Install DeepSpeed using
-
+
Starting AllTalk TTS WebUI:
+-
+
- Launch the AllTalk TTS interface with
./start_alltalk.shand enable DeepSpeed.
+
+ - Launch the AllTalk TTS interface with
Troubleshooting
+-
+
- If setting
CUDA_HOMEresults in path duplication errors (e.g., +.../bin/bin/nvcc), you can correct this by unsettingCUDA_HOMEwith +unset CUDA_HOMEand then adding the correct path to your system's PATH variable. +
+ - Always verify paths and compatibility with other CUDA-dependent applications to avoid conflicts. +
- If you have multiple versions of the Nvidia CUDA Development Toolkit installed, you will have to specify + the version number in step 8 for the CUDA_HOME path. +
- If it became necessary to uninstall DeepSpeed, you can do so by start the Python enviroment and then
+ running
pip uninstall deepspeed
+
🟢 Windows Installation
+You have 2x options for how to setup DeepSpeed on Windows. Pre-compiled wheel files for specific Python, CUDA and + Pytorch builds, or manually compiling DeepSpeed.
+Windows DeepSpeed - Pre-Compiled Wheels (Quick and Easy)
+ +DeepSpeed Installation with Pre-compiled Wheels
+-
+
-
+
Introduction to Pre-compiled Wheels:
+-
+
- The
atsetup.batutility simplifies the installation of DeepSpeed by automatically + downloading and installing pre-compiled wheel files. These files are tailored for + specific versions of Python, CUDA, and PyTorch, ensuring compatibility with + both the Standalone Installation and a standard build of + Text-generation-webui. +
+
+ - The
-
+
Manual Installation of Pre-compiled Wheels:
+-
+
- If needed, pre-compiled DeepSpeed wheel files that I have built are available on the Releases Page. You can manually
+ install or uninstall these wheels using the following commands:
-
+
- Installation:
pip install {deep-speed-wheel-file-name-here}
+ - Uninstallation:
pip uninstall deepspeed
+
+ - Installation:
+ - If needed, pre-compiled DeepSpeed wheel files that I have built are available on the Releases Page. You can manually
+ install or uninstall these wheels using the following commands:
-
+
Using
+atsetup.batfor Simplified Management:-
+
- For those running the Standalone Installation or a standard build of Text-generation-webui, the
+
atsetup.batutility offers the simplest and most efficient way to manage DeepSpeed + installations on Windows. +
+
+ - For those running the Standalone Installation or a standard build of Text-generation-webui, the
+
Windows DeepSpeed - Manual Compilation
+ +Manual DeepSpeed Wheel Compilation
+-
+
-
+
Preparation for Manual Compilation:
+-
+
- Manual compilation of DeepSpeed wheels is an advanced process that requires:
-
+
- 1-2 hours of your time for initial setup and compilation. +
- 6-10GB of disk space on your computer. +
- A solid technical understanding of Windows environments and Python. +
+
+ - Manual compilation of DeepSpeed wheels is an advanced process that requires:
-
+
Understanding Wheel Compatibility:
+-
+
- A compiled DeepSpeed wheel is uniquely tied to the specific versions of Python, PyTorch, and + CUDA used during its compilation. If any of these versions are changed, you will need to compile + a new DeepSpeed wheel to ensure compatibility. +
+ -
+
Compiling DeepSpee Resources:
+-
+
- Myself and @S95Sedan have worked to simplify the + compilation process. @S95Sedan has notably improved + the process for later versions of DeepSpeed, ensuring ease of build on Windows. +
- Because @S95Sedan is now maintaining the instructions
+ for compiling DeepSpeed on Windows, please visit @S95Sedan's
DeepSpeed GitHub page.
+
+
🔵🟢 DeepSpeed Performance Example
+
+
🆘 Support Requests, Troubleshooting & Feature + requests
+I'm thrilled to see the enthusiasm and engagement with AllTalk! Your feedback and questions are invaluable, + helping to make this project even better. To ensure everyone gets the help they need efficiently, please + consider the following before submitting a support request:
+Consult the Documentation: A comprehensive guide and FAQ sections (below) are available to help + you navigate AllTalk. Many common questions and troubleshooting steps are covered here.
+Search Past Discussions: Your issue or question might already have been addressed in the + discussions area or closed + issues. Please use the search function to see if there's an existing solution or advice that applies + to your situation.
+Bug Reports: If you've encountered what you believe is a bug, please first check the Updates & Bug Fixes List to see if it's + a known issue or one that's already been resolved. If not, I encourage you to report it by raising a bug + report in the Issues section, providing as much + detail as possible to help identify and fix the issue.
+Feature Requests: The current Feature request list can be found here. I love hearing your ideas for + new features! While I can't promise to implement every suggestion, I do consider all feedback carefully. + Please share your thoughts in the Discussions + area or via a Feature Request in the Issues + section.
+ + ++
🟨 Help with problems
+🔄 Minor updates/bug + fixes list can be found here
+🟨 How to make a diagnostics report file
+ +If you are on a Windows machine or a Linux machine, you should be able to use the atsetup.bat or
+ ./atsetup.sh utility to create a diagnositcs file. If you are unable to use the
+ atsetup utility, please follow the instructions below.
+
Manually making a diagnostics report file
+ +-
+
- Open a command prompt window and start the Python environment. Depending on your setup
+ (Text-generation-webui or Standalone AllTalk), the steps to start the Python environment vary:
+
+
-
+
-
+
For Text-generation-webui Users:
+-
+
- Navigate to the Text-generation-webui directory:
-
+
cd text-generation-webui
+
+ - Start the Python environment suitable for your OS:
-
+
- Windows:
cmd_windows.bat
+ - Linux:
./cmd_linux.sh
+ - macOS:
cmd_macos.sh
+ - WSL (Windows Subsystem for Linux):
cmd_wsl.bat
+
+ - Windows:
- Move into the AllTalk directory:
-
+
cd extensions/alltalk_tts
+
+
+ - Navigate to the Text-generation-webui directory:
-
+
For Standalone AllTalk Users:
+-
+
- Navigate to the
alltalk_ttsfolder:-
+
cd alltalk_tts
+
+ - Start the Python environment:
-
+
- Windows:
start_environment.bat
+ - Linux:
./start_environment.sh
+
+ - Windows:
+
+If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
+ - Navigate to the
-
+
-
+
Run the diagnostics and select the requirements file name you installed AllTalk with:
+-
+
python diagnostics.py
+
+ -
+
You will have an on screen output showing your environment setttings, file versions request vs + whats installed and details of your graphics card (if Nvidia). This will also create a file + called
+diagnostics.login thealltalk_ttsfolder, that you can upload + if you need to create a support ticket on here.
+
+
Installation and Setup Issues
+🟨 Understanding Python Environments Simplified
+ +Think of Python environments like different rooms in your house, each designed for a specific purpose. Just + as you wouldn't cook in the bathroom or sleep in the kitchen, different Python applications need their + own "spaces" or environments because they have unique requirements. Sometimes, these requirements + can clash with those of other applications (imagine trying to cook a meal in a bathroom!). To avoid this, + you can create separate Python environments.
+Why Separate Environments?
+Separate environments, like separate rooms, keep everything organized and prevent conflicts. For instance, + one Python application might need a specific version of a library or dependency, while another requires a + different version. Just as you wouldn't store kitchen utensils in the bathroom, you wouldn't want + these conflicting requirements to interfere with each other. Each environment is tailored and customized for + its application, ensuring it has everything it needs without disrupting others.
+How It Works in Practice:
+Standalone AllTalk Installation: When you install AllTalk standalone, it's akin to
+ adding a new room to your house specifically designed for your AllTalk activities. The setup process, using
+ the atsetup utility, constructs this custom "room" (Python environment
+ alltalk_environment) with all the necessary tools and furnishings (libraries and dependencies)
+ that AllTalk needs to function smoothly, without meddling with the rest of your "house" (computer
+ system). The AllTalk environment is started each time you run start_alltalk or
+ start_environment within the AllTalk folder.
+
Text-generation-webui Installation: Similarly, installing Text-generation-webui is like
+ setting up another specialized room. Upon installation, it automatically creates its own tailored
+ environment, equipped with everything required for text generation, ensuring a seamless and conflict-free
+ operation. The Text-generation-webui environment is started each time you run
+ start_*your-os-version* or cmd_*your-os-version* within the Text-generation-webui
+ folder.
+
Managing Environments:
+Just as you might renovate a room or bring in new furniture, you can also update or modify Python + environments as needed. Tools like Conda or venv make it easy to manage these environments, allowing you to + create, duplicate, activate, or delete them much like how you might manage different rooms in your house for + comfort and functionality.
+Once you're in the right environment, by activating it, installing or updating dependencies (the tools + and furniture of your Python application) is straightforward. Using pip, a package installer for Python, you + can easily add what you need. For example, to install all required dependencies listed in a requirements.txt + file, you'd use:
+pip install -r requirements.txt
This command tells pip to read the list of required packages and versions from the requirements.txt file and + install them in the current environment, ensuring your application has everything it needs to operate. + It's like having a shopping list for outfitting a room and ensuring you have all the right items + delivered and set up.
+Remember, just as it's important to use the right tools for tasks in different rooms of your house, + it's crucial to manage your Python environments and dependencies properly to ensure your applications + run as intended.
+How do I know if I am in a Python environment?:
+When a Python environment starts up, it changes the command prompt to show the Python environment that it + currently running within that terminal/console.
+
+
🟨 Windows & Python requirements for compiling packages (ERROR: Could not build wheels for + TTS)
+ +ERROR: Microsoft Visual C++ 14.0 or greater is required or
+ ERROR: Could not build wheels for TTS. or
+ ModuleNotFoundError: No module named 'TTS
+
Python requires that you install C++ development tools on Windows. This is detailed on the Python site here. You would need to install
+ MSVCv142 - VS 2019 C++ x64/x86 build tools and Windows 10/11 SDK from the C++
+ Build tools section.
+
You can get hold of the Community edition here the during installation, selecting
+ C++ Build tools and then MSVCv142 - VS 2019 C++ x64/x86 build tools and
+ Windows 10/11 SDK.
+
🟨 Standalone Install - start_{youros}.xx opens and closes instantly and AllTalk doesnt start
+ +This is more than likely caused by having a - in your folder path e.g.
+ c:\myfiles\alltalk_tts-main. In this circumstance you would be best renaming the folder to
+ remove the - from its name e.g. c:\myfiles\alltalk_tts, delete the
+ alltalk_environment folder and start_alltalk.bat or start_alltalk.sh
+ and then re-run atsetup to re-create the environment and startup files.
+
🟨 I think AllTalks requirements file has installed something another extension doesn't like
+ +Ive paid very close attention to not impact what Text-generation-webui is requesting on a + factory install. This is one of the requirements of submitting an extension to Text-generation-webui. If you + want to look at a comparison of a factory fresh text-generation-webui installed packages (with cuda 12.1, + though AllTalk's requirements were set on cuda 11.8) you can find that comparison here. This comparison shows that AllTalk is + requesting the same package version numbers as Text-generation-webui or even lower version numbers (meaning + AllTalk will not update them to a later version). What other extensions do, I cant really account for that. +
+I will note that the TTS engine downgrades Pandas data validator to 1.5.3 though its unlikely to cause any
+ issues. You can upgrade it back to text-generation-webui default (december 2023) with
+ pip install pandas==2.1.4 when inside of the python environment. I have noticed no ill effects
+ from it being a lower or higher version, as far as AllTalk goes. This is also the same behaviour as the
+ Coqui_tts extension that comes with Text-generation-webui.
+
Other people are reporting issues with extensions not starting with errors about Pydantic e.g.
+ pydantic.errors.PydanticImportError: BaseSettings` has been moved to the pydantic-settings package. See https://docs.pydantic.dev/2.5/migration/#basesettings-has-moved-to-pydantic-settings for more details.
+
Im not sure if the Pydantic version has been recently updated by the Text-generation-webui installer, but
+ this is nothing to do with AllTalk. The other extension you are having an issue with, need to be updated to
+ work with Pydantic 2.5.x. AllTalk was updated in mid december to work with 2.5.x. I am not specifically
+ condoning doing this, as it may have other knock on effects, but within the text-gen Python environment, you
+ can use pip install pydantic==2.5.0 or pip install pydantic==1.10.13 to change the
+ version of Pydantic installed.
🟨 I am having problems getting AllTalk to start after changing settings or making a custom setup/model + setup.
+ +I would suggest following Problems + Updating and if you still have issues after that, you can raise an issue here
+Networking and Access Issues
+🟨 I cannot access AllTalk from another machine on my Network
+ +You will need to change the IP address within AllTalk's settings from being 127.0.0.1, which only allows + access from the local machine its installed on. To do this, please see Changing + AllTalks IP address & Accessing AllTalk over your Network at the top of this page.
+You may also need to allow access through your firewall or Antivirus package to AllTalk.
+🟨 I am running a Headless system and need to change the IP Address manually as I cannot reach the + config page
+ +To do this you can edit the confignew.json file within the alltalk_tts folder. You
+ would look for "ip_address": "127.0.0.1", and change the
+ 127.0.0.1 to your chosen IP address,then save the file and start AllTalk.
+
When doing this, be careful not to impact the formatting of the JSON file. Worst case, you can re-download a
+ fresh copy of confignew.json from this website and that will put you back to a factory setting.
+
Configuration and Usage Issues
+🟨 I activated DeepSpeed in the settings page, but I didnt install DeepSpeed yet and now I have issues + starting up
+ +You can either follow the Problems
+ Updating and fresh install your config. Or you can edit the confignew.json file within
+ the alltalk_tts folder. You would look for '"deepspeed_activate": true,' and
+ change the word true to false `"deepspeed_activate": false,' ,then save the file and try
+ starting again.
If you want to use DeepSpeed, you need an Nvidia Graphics card and to install DeepSpeed on your system. + Instructions are here +
+🟨 I am having problems updating/some other issue where it wont start up/Im sure this is a bug +
+ +Please see Problems Updating. If that
+ doesnt help you can raise an ticket here. It
+ would be handy to have any log files from the console where your error is being shown. I can only losely
+ support custom built Python environments and give general pointers. Please create a
+ diagnostics.log report file to submit with a support request.
+
Also, is your text-generation-webui up to date? instructions + here
+🟨 I see some red "asyncio" messages
+ +As far as I am aware, these are to do with the chrome browser the gradio text-generation-webui in some way. I + raised an issue about this on the text-generation-webui here where you can see that + AllTalk is not loaded and the messages persist. Either way, this is more a warning than an actual issue, so + shouldnt affect any functionality of either AllTalk or text-generation-webui, they are more just an + annoyance.
+Performance and Compatibility Issues
+🟨 Warning TTS Subprocess has NOT started up yet, Will keep trying for 120 seconds maximum. Please + wait. It times out after 120 seconds.
+ When the subprocess is starting 2x things are occurring:
+ +
A) Its trying to load the voice model into your graphics card VRAM (assuming you have a
+ Nvidia Graphics card, otherwise its your system RAM)
+ B) Its trying to start up the mini-webserver and send the "ready" signal back to
+ the main process.
+
Before giving other possibilities a go, some people with old machines are finding their
+ startup times are very slow 2-3 minutes. Ive extended the allowed time within the script
+ from 1 minute to 2 minutes. If you have an older machine and wish to try extending this
+ further, you can do so by editing script.py and changing startup_wait_time = 120
+ (120 seconds, aka 2 minutes) at the top of the script.py file, to a larger value e.g
+ startup_wait_time = 240 (240 seconds aka 4 minutes).
+
Note: If you need to create a support ticket, please create a diagnostics.log
+ report file to submit with a support request. Details on doing this are above.
Other possibilities for this issue are:
+-
+
-
+
You are starting AllTalk in both your
+CMD FLAG.txtandsettings.yamlfile. + TheCMD FLAG.txtyou would have manually edited and thesettings.yamlis + the one you change and save in thesessiontab of text-generation-webui and you can +Save UI defaults to settings.yaml. Please only have one of those two starting up + AllTalk. +
+ -
+
You are not starting text-generation-webui with its normal Python environment. Please start it with + start_{your OS version} as detailed here + (
+start_windows.bat,./start_linux.sh,start_macos.shor +start_wsl.bat) OR (cmd_windows.bat,./cmd_linux.sh, +cmd_macos.shorcmd_wsl.batand thenpython server.py). +
+ -
+
You have installed the wrong version of DeepSpeed on your system, for the wrong version of + Python/Text-generation-webui. You can go to your text-generation-webui folder in a terminal/command + prompt and run the correct cmd version for your OS e.g. (
+cmd_windows.bat, +./cmd_linux.sh,cmd_macos.shorcmd_wsl.bat) and then you can + typepip uninstall deepspeedthen try loading it again. If that works, please see here + for the correct instructions for installing DeepSpeed here. +
+ -
+
You have an old version of text-generation-webui (pre Dec 2023) I have not tested on older versions + of text-generation-webui, so cannot confirm viability on older versions. For instructions on + updating the text-generation-webui, please look here + (
+update_linux.sh,update_windows.bat,update_macos.sh, or +update_wsl.bat). +
+ -
+
You already have something running on port 7851 on your computer, so the mini-webserver cant start on + that port. You can change this port number by editing the
+confignew.jsonfile and + changing"port_number": "7851"to +"port_number": "7602"or any port number you wish that isn’t + reserved. Only change the number and save the file, do not change the formatting of the document. + This will at least discount that you have something else clashing on the same port number. +
+ -
+
You have antivirus/firewalling that is blocking that port from being accessed. If you had to do + something to allow text-generation-webui through your antivirus/firewall, you will have to do that + for this too.
+
+ -
+
You have quite old graphics drivers and may need to update them.
+
+ -
+
Something within text-generation-webui is not playing nicely for some reason. You can go to your + text-generation-webui folder in a terminal/command prompt and run the correct cmd version for your + OS e.g. (
+cmd_windows.bat,./cmd_linux.sh,cmd_macos.shor +cmd_wsl.bat) and then you can typepython extensions\alltalk_tts\script.py+ and see if AllTalk starts up correctly. If it does then something else is interfering. +
+ -
+
Something else is already loaded into your VRAM or there is a crashed python process. Either check + your task manager for erroneous Python processes or restart your machine and try again.
+
+ -
+
You are running DeepSpeed on a Linux machine and although you are starting with +
+./start_linux.shAllTalk is failing there on starting. This is because + text-generation-webui will overwrite some environment variables when it loads its python + environment. To see if this is the problem, from a terminal go into your text-generation-webui + folder and./cmd_linux.shthen set your environment variable again e.g. +export CUDA_HOME=/usr/local/cuda(this may vary depending on your OS, but this is the + standard one for Linux, and assuming you have installed the CUDA toolkit), then +python server.pyand see if it starts up. If you want to edit the environment + permanently you can do so, I have not managed to write full instructions yet, but here is the conda + guide here. +
+ -
+
You have built yourself a custom Python environment and something is funky with it. This is very hard + to diagnose as its not a standard environment. You may want to updating text-generation-webui and re + installing its requirements file (whichever one you use that comes down with text-generation-webui). +
+
🟨 I have multiple GPU's and I have problems running Finetuning
+ +Finetuning pulls in various other scripts and some of those scripts can have issues with multiple Nvidia + GPU's being present. Until the people that created those other scripts fix up their code, there is a + workaround to temporarily tell your system to only use the 1x of your Nvidia GPU's. To do this:
+-
+
-
+
Windows - You will start the script with +
+set CUDA_VISIBLE_DEVICES=0 && python finetune.py
+ After you have completed training, you can reset back with +set CUDA_VISIBLE_DEVICES=
+
+ -
+
Linux - You will start the script with +
+CUDA_VISIBLE_DEVICES=0 python finetune.py
+ After you have completed training, you can reset back with +unset CUDA_VISIBLE_DEVICES
+
+
Rebooting your system will also unset this. The setting is only applied temporarily.
+Depending on which of your Nvidia GPU's is the more powerful one, you can change the 0 to
+ 1 or whichever of your GPU's is the most powerful.
+
🟨 Firefox - Streaming Audio doesnt work on Firefox
+ +This is a long standing issue with Firefox and one I am unable to resolve. The solution is to use another web + browser if you want to use Streaming audio. For details of my prior invesitigation please look at this ticket
+Application Specific Issues
+🟨 SillyTavern - I changed my IP address and now SillyTavern wont connect with AllTalk
+ +SillyTavern checks the IP address when loading extensions, saving the IP to its configuration only if the + check + succeeds. For whatever reason, SillyTavern's checks dont always allow changing its IP address a second + time.
+ +To manually change the IP address:
+-
+
- Navigate to the SillyTavern Public folder located at
/sillytavern/public/.
+ - Open the
settings.jsonfile.
+ - Look for the AllTalk section and find the
provider_endpointentry.
+ - Replace
localhostwith your desired IP address, for example,192.168.1.64. +
+
TTS Generation Issues & Questions
+🟨 XTTS - Does the XTTS AI Model Support Emotion Control or Singing?
+ +No, the XTTS AI model does not currently support direct control over emotions or singing capabilities. While + XTTS infuses generated speech with a degree of emotional intonation based on the context of the text, users + cannot explicitly control this aspect. It's worth noting that regenerating the same line of TTS may + yield slightly different emotional inflections, but there is no way to directly control it with XTTS.
+🟨 XTTS - Skips, repeats or pronunciation Issues
+ +Firstly, it's important to clarify that the development and maintenance of the XTTS AI models and core + scripts are handled by Coqui, with additional + scripts and libraries from entities like huggingface among many other Python scripts + and libraries used by AllTalk.
+AllTalk is designed to be a straightforward interface that simplifies setup and interaction with AI TTS + models like XTTS. Currently, AllTalk supports the XTTS model, with plans to include more models in the + future. Please understand that the deep inner workings of XTTS, including reasons why it may skip, repeat, + or mispronounce, along with 3rd party scripts and libraries utilized, are ultimately outside my control.
+Although I ensure the text processed through AllTalk is accurately relayed to the XTTS model speech + generation process, and I have aimed to mitigate as many issues as much as possible; skips, repeats and bad + pronounciation can still occur.
+Certain aspects I have not been able to investigate due to my own time limitations, are:
-
+
- The impact of DeepSpeed on TTS quality. Is this more likely to cause skips or repetition? +
- Comparative performance between different XTTS model versions (e.g., 2.0.3 vs. 2.0.2) regarding audio + quality and consistency. +
From my experience and anecdotally gained knowledge:
-
+
- Lower quality voice samples tend to produce more anomalies in generated speech. +
- Model finetuning with high-quality voice samples significantly reduces such issues, enhancing overall + speech quality. +
- Unused/Excessive punctuation causes issues e.g. asterisks
*, hashes#, + brackets()etc. Many of these AllTalk will filter out.
+
So for example, the female_01.wav file that is provided with AllTalk is a studio quality voice
+ sample, which the XTTS model was trained on. Typically you will find it unlikely that anomolies occur with
+ TTS generation when using this voice sample. Hence good quality samples and finetuning, generally improve
+ results with XTTS.
If you wish to try out the XTTS version 2.0.3 model and see if it works better, you can download it from here, replacing all the files within your
+ /alltalk_tts/models/xttsv2_2.0.2 folder. This is on my list to both test version 2.0.3 more,
+ but also build a more flexible TTS models downloader, that will not only accomdating other XTTS models, but
+ also other TTS engines. If you try the XTTS version 2.0.3 model and gleen any insights, please let me know.
+
+
⚫ Finetuning a model
+If you have a voice that the model doesnt quite reproduce correctly, or indeed you just want to improve the
+ reproduced voice, then finetuning is a way to train your "XTTSv2 local" model (stored in
+ /alltalk_tts/models/xxxxx/) on a specific voice. For this you will need:
-
+
- An Nvidia graphics card. (Please see the help section note + if you have multiple Nvidia GPU's). +
- 18GB of disk space free (most of this is used temporarily) +
- At least 2 minutes of good quality speech from your chosen speaker in mp3, wav or flacc format, in one or + more files (have tested as far as 20 minutes worth of audio). +
- As a side note, many people seem to think that the Whisper v2 model (used on Step 1) is giving better + results at generating training datasets, so you may prefer to try that, as opposed to the Whisper 3 model. + +
⚫ How will this work/How complicated is it?
+Everything has been done to make this as simple as possible. At its simplest, you can literally just download a + large chunk of audio from an interview, and tell the finetuning to strip through it, find spoken parts and build + your dataset. You can literally click 4 buttons, then copy a few files and you are done. At it's more + complicated end you will clean up the audio a little beforehand, but its still only 4x buttons and copying a few + files.
+⚫ The audio you will use
+I would suggest that if its in an interview format, you cut out the interviewer speaking in audacity or your + chosen audio editing package. You dont have to worry about being perfect with your cuts, the finetuning Step 1 + will go and find spoken audio and cut it out for you. Is there is music over the spoken parts, for best quality + you would cut out those parts, though its not 100% necessary. As always, try to avoid bad quality audio with + noises in it (humming sounds, hiss etc). You can try something like Audioenhancer to try clean up noisier audio. There is no need to + down-sample any of the audio, all of that is handled for you. Just give the finetuning some good quality audio + to work with.
+⚫ Can I Finetune a model more than once on + more than one voice
+Yes you can. You would do these as multiple finetuning's, but its absolutely possible and fine to do. + Finetuning the XTTS model does not restrict it to only being able to reproduce that 1x voice you trained it on. + Finetuning is generally nuding the model in a direction to learn the ability to sound a bit more like a voice + its not heard before.
+⚫ A note about anonymous training + Telemetry information & disabling it
+Portions of Coqui's TTS trainer scripts gather anonymous training information which you can disable. Their
+ statement on this is listed here. If you start
+ AllTalk Finetuning with start_finetuning.bat or ./start_finetuning.sh telemetry will
+ be disabled. If you manually want to disable it, please expand the below:
Manually disable telemetry
+ +
Before starting finetuning, run the following in your terminal/command prompt:
+-
+
- On Windows by typing
set TRAINER_TELEMETRY=0
+ - On Linux & Mac by typing
export TRAINER_TELEMETRY=0
+
Before you start finetune.py. You will now be able to finetune offline and no anonymous training
+ data will be sent.
⚫ Prerequisites for Fine-tuning + with Nvidia CUDA Development Toolkit 11.8
+All the requirements for Finetuning will be installed by using the atsetup utility and installing your correct + requirements (Standalone or for Text-generation-webui). The legacy manual instructions are stored below, however + these shouldnt be required.
+Legacy manual instructions for installing Nvidia CUDA Development Toolkit 11.8
+ - To perform fine-tuning, a specific portion of the **Nvidia CUDA Development Toolkit v11.8** must be installed. + This is crucial for step 1 of fine-tuning. The objective is to minimize the installation footprint by installing + only the essential components.
+ - The **Nvidia CUDA Development Toolkit v11.8** operates independently from your graphics card drivers and the + CUDA version utilized by your Python environment.
+ - This installation process aims to keep the download and install size as minimal as possible, however a full + install of the tookit requires 3GB's of disk space.
+ - When running Finetuning it will require upto 20GB's of temporary disk space, so please ensure you have this + space available and preferably use a SSD or NVME drive. + +
-
+
-
+
Download the Toolkit:
+-
+
- Obtain the network install version of the Nvidia CUDA Development Toolkit 11.8 + from Nvidia's + Archive. +
+ -
+
Run the Installer:
+-
+
- Choose Custom (Advanced) installation. +
- Deselect all options initially. +
- Select the following components:
-
+
CUDA>Development>Compiler> +nvcc+
+ CUDA>Development>Libraries> +CUBLAS(both development and runtime) +
+
+
+ -
+
Configure Environment Search Path:
+-
+
-
+
It's essential that
+nvccand CUDA 11.8 library files are discoverable in + your environment's search path. Adjustments can be reverted post-fine-tuning if desired. +For Windows:
+-
+
- Edit the
Pathenvironment variable to include +C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8\bin. +
+ - Add
CUDA_HOMEand set its path to +C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8.+
+
For Linux:
+-
+
- The path may vary by Linux distribution. Here's a generic setup:
-
+
-
+
+export CUDA_HOME=/usr/local/cuda
+ -
+
+export PATH=${CUDA_HOME}/bin:${PATH}
+ -
+
+export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+ -
+
Consider adding these to your
+~/.bashrcfor permanence, or apply + temporarily for the current session by running the above commands each time + you start your Python environment.
+
+ -
+
Note: If using Text-generation-webui, its best to set these temporarily.
+
+ - Edit the
+ -
+
-
+
Verify Installation:
+-
+
- Open a new terminal/command prompt to refresh the search paths. +
- In a terminal or command prompt, execute
nvcc --version.
+ - Success is indicated by a response of
Cuda compilation tools, release 11.8.+ Specifically, ensure it is version 11.8.
+
+ -
+
Troubleshooting:
+-
+
- If the correct version isn't reported, recheck your environment path settings for accuracy + and potential conflicts with other CUDA versions. +
Additional Note on Torch and Torchaudio:
+-
+
- Ensure Torch and Torchaudio are CUDA-enabled (any version), which is separate from the CUDA Toolkit
+ installation. CUDA 11.8 corresponds to
cu118and CUDA 12.1 tocu121in AllTalk + diagnostics.
+ - Failure to install CUDA for Torch and Torchaudio will result in Step 2 of fine-tuning failing. These
+ requirements are distinct from the CUDA Toolkit installation, so avoid conflating the two.
⚫ Starting Fine-tuning
+NOTE: Ensure AllTalk has been launched at least once after any updates to download necessary + files for fine-tuning.
+-
+
-
+
Close Resource-Intensive Applications:
+-
+
- Terminate any applications that are using your GPU/VRAM to ensure enough resources for fine-tuning. + +
+ -
+
Organize Voice Samples:
+-
+
- Place your audio samples into the following directory:
+
/alltalk_tts/finetune/put-voice-samples-in-here/+
+
+ - Place your audio samples into the following directory:
+
Depending on your setup (Text-generation-webui or Standalone AllTalk), the steps to start the Python environment + vary:
+-
+
-
+
For Standalone AllTalk Users:
+-
+
- Navigate to the
alltalk_ttsfolder:-
+
cd alltalk_tts
+
+ - Start the Python environment:
-
+
- Windows:
start_finetune.bat
+ - Linux:
./start_finetune.sh
+
+ - Windows:
+ - Navigate to the
-
+
For Text-generation-webui Users:
+-
+
- Navigate to the Text-generation-webui directory:
-
+
cd text-generation-webui
+
+ - Start the Python environment suitable for your OS:
-
+
- Windows:
cmd_windows.bat
+ - Linux:
./cmd_linux.sh
+ - macOS:
cmd_macos.sh
+ - WSL (Windows Subsystem for Linux):
cmd_wsl.bat
+
+ - Windows:
- Move into the AllTalk directory:
-
+
cd extensions/alltalk_tts
+
+ - Linux users only need to run this command:
-
+
export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'`
+
+ - Start the fine-tuning process with the command:
-
+
python finetune.py
+
+
+
+If you're unfamiliar with Python environments and wish to learn more, consider reviewing + Understanding Python Environments Simplified in the Help section. +
+
+ - Navigate to the Text-generation-webui directory:
-
+
-
+
Pre-Flight Checklist:
+-
+
- Go through the pre-flight checklist to ensure readiness. Address any issues flagged as + "Fail". +
+ -
+
Post Fine-tuning Actions:
+-
+
- Upon completing fine-tuning, the final tab will guide you on managing your files and relocating your + newly trained model to the appropriate directory. +
+
These steps guide you through the initial preparations, starting the Python environment based on your setup, and + the fine-tuning process itself. Ensure all prerequisites are met to facilitate a smooth fine-tuning experience. +
+⚫ How many Epochs etc is the right amount?
+In finetuning the suggested/recommended amount of epochs, batch size, evaluation percent etc is already set. + However, there is no absolutely correct answer to what the settings should be, it all depends on what you are + doing.
+-
+
- If you just want to train a normal human voice that is in an existing language, for most people’s needs, the + base settings would work fine. You may choose to increase the epochs up to maybe 20, or run a second round + of training if needed. +
- If you were training an entirely new language, you would need a huge amount of training data and it requires + around 1000 epochs (based on things I can find around the internet of people who tried this). +
- If you are training a cartoon style voice in an existing language, it may need well upwards of 40 epochs + until it can reproduce that voice with some success. +
There are no absolute correct settings, as there are too many variables, ranging from the amount of samples you + are using (5 minutes worth? 4 hours worth? etc), if they are similar samples to what the AI model already + understands, so on and so forth. Coqui whom originally trained the model usually say something along the lines + of, once you’ve trained it X amount, if it sounds good then you are done and if it doesn’t, train it more.
+⚫ Evaluation Data Percentage
+In the process of finetuning, it's crucial to balance the data used for training the model against the data
+ reserved for evaluating its performance. Typically, a portion of the dataset is set aside as an 'evaluation
+ set' to assess the model's capabilities in dealing with unseen data. On Step 1 of finetuning you have
+ the option to adjust this evaluation data percentage, offering more control over your model training
+ process.
+ Why Adjust the Evaluation Percentage?
+ Adjusting the evaluation percentage can be beneficial in scenarios with limited voice samples.
+ When dealing with a smaller dataset, allocating a slightly larger portion to training could enhance the
+ model's ability to learn from these scarce samples. Conversely, with abundant data, a higher evaluation
+ percentage might be more appropriate to rigorously test the model's performance. There are currently no
+ absolutely optimal split percentages as it varies by dataset.
+
-
+
-
+
Default Setting: The default evaluation percentage is set at 15%, which is a balanced + choice for most datasets.
+
+ -
+
Adjustable Range: Users can now adjust this percentage, but it’s generally recommend + keeping it between 5% and 30%.
+-
+
- Lower Bound: A minimum of 5% ensures that there's enough data to evaluate model + performance. +
- Upper Bound: Its suggested not exceeding 30% for evaluation to avoid limiting the + amount of data available for training. +
+ -
+
Understanding the Impact: Before adjusting this setting, it's important to + understand its impact on model training and evaluation. Incorrect adjustments can lead to suboptimal + model performance.
+
+ -
+
Gradual Adjustments: For those unfamiliar with the process, we recommend reading up on + training data and training sets, then making small, incremental changes and observing their effects.
+
+ -
+
Data Quality: Regardless of the split, the quality of the audio data is paramount. + Ensure that your datasets are built from good quality audio with enough data within them.
+
+
⚫ Using a Finetuned model in Text-generation-webui
+At the end of the finetune process, you will have an option to
+ Compact and move model to /trainedmodel/ this will compact the raw training file and move it to
+ /model/trainedmodel/. When AllTalk starts up within Text-generation-webui, if it finds a model in
+ this location a new loader will appear in the interface for XTTSv2 FT and you can use this to load
+ your finetuned model.
Be careful not to train a new model from the base model, then
+ overwrite your current /model/trainedmodel/ if you want a seperately trained
+ model. This is why there is an OPTION B to move your just trained model to
+ /models/lastfinetuned/.
+
⚫ Training one model with multiple voices
+At the end of the finetune process, you will have an option to
+ Compact and move model to /trainedmodel/ this will compact the raw training file and move it to
+ /model/trainedmodel/. This model will become available when you start up finetuning. You will have
+ a choice to train the Base Model or the Existing finetuned model (which is the one in
+ /model/trainedmodel/). So you can use this to keep further training this model with additional
+ voices, then copying it back to /model/trainedmodel/ at the end of training.
+
⚫ Do I need to keep the raw training data/model?
+If you've compacted and moved your model, its highly unlikely you would want to keep that data, however the + choice is there to keep it if you wish. It will be between 5-10GB in size, so most people will want to delete + it.
+ ++
⬜ AllTalk TTS Generator
+ + +AllTalk TTS Generator is the solution for converting large volumes of text into speech using the voice of your
+ choice. Whether you're creating audio content or just want to hear text read aloud, the TTS Generator is
+ equipped to handle it all efficiently. Please see here for a quick demo
The link to open the TTS generator can be
+ found on the built-in Settings and Documentation page.
DeepSpeed is
+ highly recommended to speed up generation. Low VRAM would be best turned off
+ and your LLM model unloaded from your GPU VRAM (unload your model). No Playback will reduce
+ memory overhead on very large generations (15,000 words or more). Splitting Export to Wav into
+ smaller groups will also reduce memory overhead at the point of exporting your wav files (so good for low memory
+ systems).
+
⬜ Estimated Throughput
+This will vary by system for a multitude of reasons, however, while generating a 58,000 word document to TTS, + with DeepSpeed enabled, LowVram disabled, splitting size 2 and on an Nvidia RTX 4070, throughput was around + 1,000 words per minute. Meaning, this took 1 hour to generate the TTS. Exporting to combined wavs took about 2-3 + minutes total.
+⬜ Quick Start
+-
+
- Text Input: Enter the text you wish to convert into speech in the 'Text Input' box. + +
- Generate TTS: Hit this to start the text-to-speech conversion. +
- Pause/Resume: Used to pause and resume the playback of the initial generation of wavs or + the stream. +
- Stop Playback: This will stop the current audio playing back. It does not stop the text + from being generated however. + Once you have sent text off to be generated, either as a stream or wav file generation, the TTS server will + remain busy until this process has competed. As such, think carefully as to how much you want to send to the + server. + If you are generating wav files and populating the queue, you can generate one lot of text to speech, then + input your next lot of text and it will continue adding to the list. +
⬜ Customization and Preferences
+-
+
- Character Voice: Choose the voice that will read your text. +
- Language: Select the language of your text. +
- Chunk Sizes: Decide the size of text chunks for generation. Smaller sizes are recommended + for better TTS quality. +
⬜ Interface and Accessibility
+-
+
- Dark/Light Mode: Switch between themes for your visual comfort. +
- Word Count and Generation Queue: Keep track of the word count and the generation progress. + +
⬜ TTS Generation Modes
+-
+
- Wav Chunks: Perfect for creating audio books, or anything you want to keep long term. + Breaks down your text into manageable wav files and queues them up. Generation begins automatically, and + playback will start after a few chunks have been prepared ahead. You can set the volume to 0 if you don’t + want to hear playback. With Wav chunks, you can edit and/or regenerate portions of the TTS as needed. +
- Streaming: For immediate playback without the ability to save. Ideal for on-the-fly speech
+ generation and listening. This will not generate wav files and it will play back through your browser. You
+ cannot stop the server generating the TTS once it has been sent.
+ With wav chunks you can either playback “In Browser” which is the web page you are on, or “On Server” which + is through the console/terminal where AllTalk is running from, or "No Playback". Only generation + “In Browser” can play back smoothly and populate the Generated TTS List. Setting the Volume will affect the + volume level played back both “In Browser” and “On Server”.
+ For generating large amounts of TTS, it's recommended to select the No + Playback option. This setting minimizes the memory usage in your web browser by avoiding the + loading and playing of audio files directly within the browser, which is particularly beneficial for + handling extensive audio generations. The definition of large will vary depending on your system RAM + availability (will update when I have more information as to guidelines). Once the audio is generated, you + can export your list to JSON (for safety) and use the Play List option to play back your + audio.
+
⬜ Playback and List Management
+-
+
- Playback Controls: Utilize 'Play List' to start from the beginning or 'Stop + Playback' to halt at any time. +
- Custom Start: Jump into your list at a specific ID to hear a particular section. +
- Regeneration and Editing: If a chunk isn't quite right, you can opt to regenerate it or + edit the text directly. Click off the text to save changes and hit regenerate for the specific line. +
- Export/Import List: Save your TTS list as a JSON file or import one. Note: Existing wav + files are needed for playback. Exporting is handy if you want to take your files away into another program + and have a list of which wav is which, or if you keep your audio files, but want to come back at a later + date, edit one or two lines, regenerate the speech and re-combine the wav’s into one new long wav. +
⬜ Exporting Your Audio
+-
+
- Export to WAV: Combine all generated TTS from the list, into one single WAV file for easy
+ download and distribution. Its always recommended to export your list to a JSON before exporting, so that
+ you have a backup, should something go wrong. You can simply re-import the list and try exporting
+ again.
When exporting, there is a file size limit of 1GB and as such you have the option to choose + how many files to include in each block of audio exported. 600 is just on the limit of 1GB, depending on the + average file size, so 500 or less is a good amount to work with. You can combine the generated files after + if you wish, in Audacity or similar.
Additionally, lower export batches will lower the memory + requirements, so if your system is low on memory (maybe 8 or 16GB system), you can use smaller export + batches to keep the memory requirement down.
+
⬜ Exporting Subtitles (SRT file)
+-
+
- Export SRT: This will scan through all wav files in your list and generate a subtitles file + that will match your exported wav file. +
⬜ Analyzing generated TTS for errors
+-
+
- Analyze TTS: This will scan through all wav files comparing each ID's orignal text with
+ the TTS generated for that ID and then flag up inconsistences. Its important to understand this is a
+ best effort process and not 100% perfect, for example:
+-
+
- Your text may have the word
theirand the automated routine that listens to your + generated TTS interprets the word asthere, aka a spelling difference.
+ - Your text may have
Examples are:(note the colon) and the automated routine that + listens to your generated TTS interprets the word as "Examples are` (note no colon as you + cannot sound out a colon in TTS), aka a punctuation difference.
+ - Your text may have
There are 100 itemsand the automated routine that listens to your + generated TTS interprets the word asThere are one hundred items, aka numbers vs the + number written out in words.
+ - There will be other examples such as double quotes. As I say, please remember this is a best
+ effort to help you identify issues.
+ - Your text may have the word
As such, there is a % Accuracy setting. This uses a couple of methods to try find things that are
+ similar e.g. taking the their and there example from above, it would identify that
+ they both sound the same, so even if the text says their and the AI listening to the generated TTS
+ interprets the word as there, it will realise that both sound the same/are similar so there is no
+ need to flag that as an error. However, there are limits to this and some things may slip through or get picked
+ up when you would prefer them not to be flagged.
The higher the accuracy you choose, the more things it will flag up, however you may get more unwanted + detections. The lower the less detections. Based on my few tests, accuracy settings between 96 to 98 seem to + generally give the best results. However, I would highly recommend you test out a small 10-20 line text and test + out the Analyze TTS button to get a feel for how it responds to different settings, as well as + things it flags up.
+You will be able to see the ID's and Text (orignal and as interpreted) by looking at the terminal/command + prompt window.
+The Analyze TTS feature uses the Whisper Larger-v2 AI engine, which will download on first use if necessary. This + will require about 2.5GB's of disk space and could take a few minutes to download, depending on your + internet connection.
+⬜ Tricks to get the model to say things correctly
+Sometimes the AI model won’t say something the way that you want it to. It could be because it’s a new word, an + acronym or just something it’s not good at for whatever reason. There are some tricks you can use to improve the + chances of it saying something correctly.
+Adding pauses
+ You can use semi-colons ";" and colons ":" to create a pause, similar to a period
+ "." which can be helpful with some splitting issues.
Acronyms
+ Not all acronyms are going to be pronounced correctly. Let’s work with the word ChatGPT. We know it
+ is pronounced "Chat G P T" but when presented to the model, it doesn’t know how to break
+ it down correctly. So, there are a few ways we could get it to break out "Chat" and the G P and T.
+ e.g.
Chat G P T.
+ Chat G,P,T.
+ Chat G.P.T.
+ Chat G-P-T.
+ Chat gee pee tea
+
All bar the last one are using ways within the English language to split out "Chat" into one word being + pronounced and then split the G, P and T into individual letters. The final example, which is to use Phonetics + will sound perfectly fine, but clearly would look wrong as far as human readable text goes. The phonetics method + is very useful in edge cases where pronunciation difficult.
+⬜ Notes on Usage
+-
+
- For seamless TTS generation, it's advised to keep text chunks under 250 characters, which you can + control with the Chunk sizes. +
- Generated audio can be played back from the list, which also highlights the currently playing chunk. +
- The TTS Generator remembers your settings, so you can pick up where you left off even after refreshing the + page. +
+
🟠 API Suite and JSON-CURL
+🟠Overview
+The Text-to-Speech (TTS) Generation API allows you to generate speech from text input using various configuration + options. This API supports both character and narrator voices, providing flexibility for creating dynamic and + engaging audio content.
+🟠 Ready Endpoint
+ Check if the Text-to-Speech (TTS) service is ready to accept requests.
+-
+
-
+
URL:
+http://127.0.0.1:7851/api/ready
- Method:GET
+curl -X GET "http://127.0.0.1:7851/api/ready"Response:
+Ready
+
🟠 Voices List Endpoint
+ Retrieve a list of available voices for generating speech.
+-
+
-
+
URL:
+http://127.0.0.1:7851/api/voices
- Method:GET
+curl -X GET "http://127.0.0.1:7851/api/voices"JSON return: +
+{"voices": ["voice1.wav", "voice2.wav", "voice3.wav"]}+
+
🟠 Current Settings Endpoint
+ Retrieve a list of available voices for generating speech.
+-
+
-
+
URL:
+http://127.0.0.1:7851/api/currentsettings
- Method:GET
+curl -X GET "http://127.0.0.1:7851/api/currentsettings"JSON return: +
+{"models_available":[{"name":"Coqui","model_name":"API TTS"},{"name":"Coqui","model_name":"API Local"},{"name":"Coqui","model_name":"XTTSv2 Local"}],"current_model_loaded":"XTTSv2 Local","deepspeed_available":true,"deepspeed_status":true,"low_vram_status":true,"finetuned_model":false}+
+name & model_name= listing the currently available models.
+current_model_loaded= what model is currently loaded into VRAM.
+deepspeed_available= was DeepSpeed detected on startup and available to be activated.
+deepspeed_status= If DeepSpeed was detected, is it currently activated.
+low_vram_status= Is Low VRAM currently enabled.
+finetuned_model= Was a finetuned model detected. (XTTSv2 FT).
+
+
🟠 Preview Voice Endpoint
+Generate a preview of a specified voice with hardcoded settings.
+-
+
-
+
URL:
+http://127.0.0.1:7851/api/previewvoice/
- Method:POST
- + Content-Type:application/x-www-form-urlencoded
+curl -X POST "http://127.0.0.1:7851/api/previewvoice/" -F "voice=female_01.wav"+Replace
+female_01.wavwith the name of the voice sample you want to hear.JSON return: +
+{"status": "generate-success", "output_file_path": "/path/to/outputs/api_preview_voice.wav", "output_file_url": "http://127.0.0.1:7851/audio/api_preview_voice.wav"}+
+
🟠 Switching Model Endpoint
+ -
+
-
+
URL:
+http://127.0.0.1:7851/api/reload
- Method:POST
+curl -X POST "http://127.0.0.1:7851/api/reload?tts_method=API%20Local"
+curl -X POST "http://127.0.0.1:7851/api/reload?tts_method=API%20TTS"
+curl -X POST "http://127.0.0.1:7851/api/reload?tts_method=XTTSv2%20Local"
+Switch between the 3 models respectively.
+
+curl -X POST "http://127.0.0.1:7851/api/reload?tts_method=XTTSv2%20FT"If you have a finetuned model in
+/models/trainedmodel/(will error otherwise)JSON return
+{"status": "model-success"}
+
🟠 Switch DeepSpeed Endpoint
+ -
+
-
+
URL:
+http://127.0.0.1:7851/api/deepspeed
- Method:POST
+curl -X POST "http://127.0.0.1:7851/api/deepspeed?new_deepspeed_value=True"+Replace True with False to disable DeepSpeed mode.
+JSON return
+{"status": "deepspeed-success"}
+
🟠 Switching Low VRAM Endpoint
+ -
+
-
+
URL:
+http://127.0.0.1:7851/api/lowvramsetting
- Method:POST
+curl -X POST "http://127.0.0.1:7851/api/lowvramsetting?new_low_vram_value=True"+Replace True with False to disable Low VRAM mode.
+JSON return
+{"status": "lowvram-success"}
+
🟠 TTS Generation Endpoint (Standard Generation)
+Streaming endpoint details are further down the page.
+-
+
- URL:
http://127.0.0.1:7851/api/tts-generate
- Method:POST
- Content-Type: +application/x-www-form-urlencoded
+
+
🟠 Example command lines (Standard Generation)
+Standard TTS generation supports Narration and will generate a wav file/blob. Standard TTS speech Example
+ (standard text) generating a time-stamped file
curl -X POST "http://127.0.0.1:7851/api/tts-generate" -d "text_input=All of this is text spoken by the character. This is text not inside quotes, though that doesnt matter in the slightest" -d "text_filtering=standard" -d "character_voice_gen=female_01.wav" -d "narrator_enabled=false" -d "narrator_voice_gen=male_01.wav" -d "text_not_inside=character" -d "language=en" -d "output_file_name=myoutputfile" -d "output_file_timestamp=true" -d "autoplay=true" -d "autoplay_volume=0.8"
+
Narrator Example (standard text) generating a time-stamped file
+curl -X POST "http://127.0.0.1:7851/api/tts-generate" -d "text_input=*This is text spoken by the narrator* \"This is text spoken by the character\". This is text not inside quotes." -d "text_filtering=standard" -d "character_voice_gen=female_01.wav" -d "narrator_enabled=true" -d "narrator_voice_gen=male_01.wav" -d "text_not_inside=character" -d "language=en" -d "output_file_name=myoutputfile" -d "output_file_timestamp=true" -d "autoplay=true" -d "autoplay_volume=0.8"
+
Note that if your text that needs to be generated contains double quotes you will need to escape them with
+ \" (Please see the narrator example).
+
🟠 Request Parameters
+🟠 text_input: The text you want the TTS engine to produce. Use escaped double quotes for + character speech and asterisks for narrator speech if using the narrator function. Example:
+-d "text_input=*This is text spoken by the narrator* \"This is text spoken by the character\". This is text not inside quotes."
+
🟠 text_filtering: Filter for text. Options:
+-
+
- none No filtering. Whatever is sent will go over to the TTS engine as raw text, which may
+ result in some odd sounds with some special characters.
- standard Human-readable text and a basic level of filtering, just to clean up some special
+ characters.
- html HTML content. Where you are using HTML entity's like "
-d "text_filtering=none"
+ -d "text_filtering=standard"
+ -d "text_filtering=html"
+
Example:
+-
+
- Standard Example:
+
*This is text spoken by the narrator* "This is text spoken by the character" This is text not inside quotes.
+
+ - HTML Example:
+
*This is text spoken by the narrator* "This is text spoken by the character" This is text not inside quotes.
+
+ - None:
Will just pass whatever characters/text you send at it.
+
🟠 character_voice_gen: The WAV file name for the character's voice.
-d "character_voice_gen=female_01.wav"
🟠 narrator_enabled: Enable or disable the narrator function. If true, minimum text filtering is + set to standard. Anything between double quotes is considered the character's speech, and anything between + asterisks is considered the narrator's speech.
+-d "narrator_enabled=true"
+ -d "narrator_enabled=false"
+
🟠 narrator_voice_gen: The WAV file name for the narrator's voice.
+-d "narrator_voice_gen=male_01.wav"
🟠 text_not_inside: Specify the handling of lines not inside double quotes or asterisks, for the + narrator feature. Options:
+-
+
- character: Treat as character speech.
- narrator: Treat as narrator speech.
-d "text_not_inside=character"
+ -d "text_not_inside=narrator"
+
🟠 language: Choose the language for TTS. Options:
+ar Arabic
+ zh-cn Chinese (Simplified)
+ cs Czech
+ nl Dutch
+ en English
+ fr French
+ de German
+ hu Hungarian
+ it Italian
+ ja Japanese
+ ko Korean
+ pl Polish
+ pt Portuguese
+ ru Russian
+ es Spanish
+ tr Turkish
+
-d "language=en"
🟠 output_file_name: The name of the output file (excluding the .wav extension).
+-d "output_file_name=myoutputfile"
🟠 output_file_timestamp: Add a timestamp to the output file name. If true, each file will have + a unique timestamp; otherwise, the same file name will be overwritten each time you generate TTS.
+-d "output_file_timestamp=true"
+ -d "output_file_timestamp=false"
+
🟠 autoplay: Enable or disable playing the generated TTS to your standard sound output device at + time of TTS generation.
+-d "autoplay=true"
+ -d "autoplay=false"
+
🟠 autoplay_volume: Set the autoplay volume. Should be between 0.1 and 1.0. Needs to be + specified in the JSON request even if autoplay is false.
+-d "autoplay_volume=0.8"
🟠 TTS Generation Response
+The API returns a JSON object with the following properties:
+-
+
- status Indicates whether the generation was successful (generate-success) or failed
+ (generate-failure).
- output_file_path The on-disk location of the generated WAV file.
- output_file_url The HTTP location for accessing the generated WAV file for browser
+ playback.
- output_cache_url The HTTP location for accessing the generated WAV file as a pushed
+ download.
Example JSON TTS Generation Response:
+{"status":"generate-success","output_file_path":"C:\\text-generation-webui\\extensions\\alltalk_tts\\outputs\\myoutputfile_1704141936.wav", "output_file_url":"http://127.0.0.1:7851/audio/myoutputfile_1704141936.wav", "output_cache_url":"http://127.0.0.1:7851/audiocache/myoutputfile_1704141936.wav"}
+
🟠 TTS Generation Endpoint (Streaming Generation)
+Streaming TTS generation does NOT support Narration and will generate an audio stream. Streaming TTS speech
+ JavaScript Example:
-
+
- URL:
http://localhost:7851/api/tts-generate-streaming
- Method:POST
- + Content-Type:application/x-www-form-urlencoded
+
// Example parameters
+ const text = "Here is some text";
+ const voice = "female_01.wav";
+ const language = "en";
+ const outputFile = "stream_output.wav";
+ // Encode the text for URL
+ const encodedText = encodeURIComponent(text);
+ // Create the streaming URL
+ const streamingUrl = `http://localhost:7851/api/tts-generate-streaming?text=${encodedText}&voice=${voice}&language=${language}&output_file=${outputFile}`;
+ // Create and play the audio element
+ const audioElement = new Audio(streamingUrl);
+ audioElement.play(); // Play the audio stream directly
+ -
+
- Text (text): This is the actual text you want to convert to speech. It should be a string
+ and must be URL-encoded to ensure that special characters (like spaces and punctuation) are correctly
+ transmitted in the URL. Example:
Hello WorldbecomesHello%20Worldwhen + URL-encoded.
+ - Voice (voice): This parameter specifies the voice type to be used for the TTS. The value
+ should match one of the available voice options in AllTalks voices folder. This is a string representing the
+ file, like
female_01.wav.
+ - Language (language): This setting determines the language in which the text should be
+ spoken. A two-letter language code (like
enfor English,frfor French, etc.).
+
+ - Output File (output_file): This parameter names the output file where the audio will be
+ streamed. It should be a string representing the file name, such as
stream_output.wav. AllTalk + will not save this as a file in its outputs folder.
+
+
Thanks & References
+Coqui TTS Engine
+ + +Extension coded by
+ + +Thanks to & Text generation webUI
+-
+
- Ooobabooga + GitHub Repository (Portions of orginal Coquii_TTS extension) +
Thanks to
+-
+
- daswer123 GitHub Profile + (Assistance with cuda to cpu moving) +
- S95Sedan GitHub Profile + (Editing the Microsoft DeepSpeed v11.x installation files so they work) +
- kanttouchthis GitHub + Profile (Portions of orginal Coquii_TTS extension) +
- Wuzzooy GitHub Profile (Trying + out the code while in development) +