diff --git a/googlecolab.ipynb b/googlecolab.ipynb index 843ae7a..408e61a 100644 --- a/googlecolab.ipynb +++ b/googlecolab.ipynb @@ -21,14 +21,14 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 9, "metadata": { "cellView": "form", "colab": { "base_uri": "https://localhost:8080/" }, "id": "W6ls-RHTQeZv", - "outputId": "a20360b3-a5cb-48c5-b885-db89e8c143f0" + "outputId": "25c62fb3-26ef-4493-8647-1d8e49bf5f01" }, "outputs": [ { @@ -44,8 +44,9 @@ } ], "source": [ - "#@markdown Mounting your Google Drive allows you to drag and drop samples/models via the `drive/Mydrive` path. This allows you
\n", - "#@markdown to store or use specific audio samples or finetuned models.

\n", + "#@markdown Click the `Play button` to the left of this message to install the requirements.

\n", + "#@markdown **OPTIONAL** Mounting your Google Drive allows you to drag and drop samples/models via the `drive/Mydrive` path. This allows you
\n", + "#@markdown to store or use specific audio samples or finetuned models.
\n", "#@markdown **Audio samples** need to be placed in `alltalk_tts/voices`
\n", "#@markdown **XTTS models** need to be placed in `alltalk_tts/models/xtts/{yourmodelfolderhere}`
\n", "mount_gdrive = False #@param{type:\"boolean\"}\n", @@ -118,110 +119,23 @@ }, "source": [ "\n", - "# Start AllTalk" + "# Start AllTalk TTS Server\n", + "\n", + "This will start the AllTalk API and Gradio Web interface. From here you can download models, generate TTS and use external applications via the API address.\n", + "\n", + "The AllTalk API address is what you would use in Kobold, SillyTavern, TGWUI's Remote extension etc if you want to generate TTS with those applications." ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "metadata": { "cellView": "form", - "id": "9PCQhSoiWvhL", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "20c6e4b0-7075-4d7c-f703-3b90e373414f" + "id": "9PCQhSoiWvhL" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "[AllTalk TTS]\u001b[94m _ _ _ \u001b[1;35m_____ _ _ \u001b[0m _____ _____ ____ \n", - "[AllTalk TTS]\u001b[94m / \\ | | |\u001b[1;35m_ _|_ _| | | __ \u001b[0m |_ _|_ _/ ___| \n", - "[AllTalk TTS]\u001b[94m / _ \\ | | |\u001b[1;35m | |/ _` | | |/ / \u001b[0m | | | | \\___ \\ \n", - "[AllTalk TTS]\u001b[94m / ___ \\| | |\u001b[1;35m | | (_| | | < \u001b[0m | | | | ___) |\n", - "[AllTalk TTS]\u001b[94m /_/ \\_\\_|_|\u001b[1;35m |_|\\__,_|_|_|\\_\\ \u001b[0m |_| |_| |____/ \n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[92mConfig file update: \u001b[93mNo Updates required\u001b[0m\n", - "[AllTalk TTS] \u001b[92mStart-up Mode : \u001b[93mStandalone mode\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[92mThis is the first time startup. Please download a start TTS model.\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[94mAvailable First Time Start-up models:\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[93m1. \u001b[94mpiper - piper\u001b[0m\n", - "[AllTalk TTS] \u001b[93m2. \u001b[94mvits - tts_models--en--vctk--vits\u001b[0m\n", - "[AllTalk TTS] \u001b[93m3. \u001b[94mxtts - xttsv2_2.0.3\u001b[0m\n", - "[AllTalk TTS] \u001b[93m4.\u001b[94m I have my own models already\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[94mIn \u001b[91m60 seconds\u001b[0m \u001b[94ma Piper model will be \u001b[91mdownloaded automatically.\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[92mEnter your choice 1-4: \u001b[0m \n", - "[AllTalk TTS]\n", - "[AllTalk TTS] No input received. Proceeding with the default model (piper).\n", - "/content/alltalk_tts/models/piper/en_US-ljspeech-high.onnx: 100% 114M/114M [00:01<00:00, 80.8MiB/s]\n", - "/content/alltalk_tts/models/piper/en_US-ljspeech-high.onnx.json: 100% 4.97k/4.97k [00:00<00:00, 14.9MiB/s]\n", - "[AllTalk TTS] piper model downloaded and configuration updated successfully.\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[93mIf you have you have UPGRADED from v1 ensure you have re-installed\u001b[0m\n", - "[AllTalk TTS] \u001b[93mthe requirements. Otherwise you will get failures and errors!\u001b[0m\n", - "[AllTalk TTS] \u001b[93mOn Linux ignore the \u001b[0m'sparse_attn requires a torch version' \u001b[93mand\u001b[0m\n", - "[AllTalk TTS] \u001b[0m'using untested triton version' \u001b[93mmessages.\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[92mWAV file deletion :\u001b[93m Disabled\u001b[0m\n", - "[AllTalk TTS] \u001b[92mGithub updated :\u001b[93m 1st July 2024 at 08:57\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[94mGoogle Colab Detected\u001b[00m\n", - "[AllTalk TTS]\n", - "[AllTalk ENG] \u001b[92mTranscoding :\u001b[93m ffmpeg found\u001b[0m\n", - "\u001b[93m [WARNING] \u001b[0m Please specify the CUTLASS repo directory as environment variable $CUTLASS_PATH\n", - "\u001b[93m [WARNING] \u001b[0m NVIDIA Inference is only supported on Ampere and newer architectures\n", - "\u001b[93m [WARNING] \u001b[0m sparse_attn requires a torch version >= 1.5 and < 2.0 but detected 2.3\n", - "\u001b[93m [WARNING] \u001b[0m using untested triton version (2.3.1), only 1.0.0 is known to be compatible\n", - "[AllTalk ENG] \u001b[92mDeepSpeed version :\u001b[93m 0.14.4 \u001b[0m\n", - "[AllTalk ENG] \u001b[92mPython Version :\u001b[93m 3.10.12\u001b[0m\n", - "[AllTalk ENG] \u001b[92mPyTorch Version :\u001b[93m 2.3.1+cu121\u001b[0m\n", - "[AllTalk ENG] \u001b[92mCUDA Version :\u001b[93m 12.1\u001b[0m\n", - "[AllTalk ENG]\n", - "[AllTalk ENG]\u001b[94m Model/Engine :\u001b[93m Piper\u001b[94m Ready\u001b[0m\n", - "[AllTalk ENG] \u001b[94mLoad time :\u001b[93m 0.00 seconds.\u001b[0m\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] \u001b[94mAPI Address :\u001b[00m \u001b[92mhttps://refresh-jackson-firefox-elementary.trycloudflare.com\u001b[00m\n", - "[AllTalk TTS] \u001b[94mGradio Light:\u001b[00m \u001b[92mhttps://lawsuit-effectively-assessing-z.trycloudflare.com\u001b[00m\n", - "[AllTalk TTS] \u001b[94mGradio Dark :\u001b[00m \u001b[92mhttps://lawsuit-effectively-assessing-z.trycloudflare.com?__theme=dark\u001b[00m\n", - "[AllTalk TTS]\n", - "themes/theme_schema@0.0.1.json: 100% 13.1k/13.1k [00:00<00:00, 45.7MB/s]\n", - "[AllTalk TTS] Please use \u001b[91mCtrl+C\u001b[0m when exiting AllTalk otherwise a\n", - "[AllTalk TTS] subprocess may continue running in the background.\n", - "[AllTalk TTS]\n", - "[AllTalk TTS] AllTalk Server Ready\n", - "Downloading LICENSE.txt...\n", - "100% 4.01k/4.01k [00:00<00:00, 11.6MiB/s]\n", - "Downloading README.md...\n", - "100% 4.26k/4.26k [00:00<00:00, 13.0MiB/s]\n", - "Downloading config.json...\n", - "100% 4.37k/4.37k [00:00<00:00, 12.1MiB/s]\n", - "Downloading model.pth...\n", - "100% 1.87G/1.87G [00:30<00:00, 61.2MiB/s]\n", - "Downloading dvae.pth...\n", - "100% 211M/211M [00:04<00:00, 50.7MiB/s]\n", - "Downloading mel_stats.pth...\n", - "100% 1.07k/1.07k [00:00<00:00, 2.90MiB/s]\n", - "Downloading speakers_xtts.pth...\n", - "100% 7.75M/7.75M [00:00<00:00, 77.9MiB/s]\n", - "Downloading vocab.json...\n", - "100% 361k/361k [00:00<00:00, 26.4MiB/s]\n", - "[AllTalk Shutdown] \u001b[94mReceived Ctrl+C, terminating subprocess. Kill your Python processes if this fails to exit.\u001b[92m\n", - "[AllTalk Shutdown] \u001b[94mReceived Ctrl+C, terminating subprocess. Kill your Python processes if this fails to exit.\u001b[92m\n", - "^C\n" - ] - } - ], + "outputs": [], "source": [ - "#@markdown Links to the API and Gradio interface will be shown when AllTalk has loaded in.
\n", - "#@markdown The API address is what you would use in SillyTavern.
\n", - "#@markdown Do not start this if you want to run Finetuning.\n", + "#@markdown Click the `Play button` to the left of this message to start AllTalk API and Gradio Interface
\n", "import re\n", "import time\n", "import json\n", @@ -296,7 +210,13 @@ { "cell_type": "markdown", "source": [ - "# Start Finetuning" + "# Start XTTS model Finetuning\n", + "\n", + "Starts the Finetuning application for XTTS models.\n", + "\n", + "You can either run the `Start AllTalk` to download the base XTTS model(s) for finetuning. Or you can use the folder icon on the left hand side of the screen and upload an XTTS model that you want to finetune. If you are manually uploading a model, you would place your model files in `models/xtts/{yourmodelfolderhere}` and you will need all the models files in that folder `config.json, dvae.pth, mel_stats.pth, model.pth, speakers_xtts.pth, vocab.json`. Without 1x model available, Finetuning will not start.\n", + "\n", + "Likewise you can download your finetuned model from there, OR copy it to your Google Drive after finetuning, for later use in AllTalk. To access the Finetuning Gradio interface, connect to the `Google Colab Finetuning url` when Finetuning has started." ], "metadata": { "id": "gDzNPMwM7qvs" @@ -305,13 +225,7 @@ { "cell_type": "code", "source": [ - "#@markdown Starts the Finetuning application for XTTS models.

\n", - "#@markdown You can either run the `Start AllTalk` to download the base XTTS model(s) for finetuning. Or you can use the folder icon on the left hand
\n", - "#@markdown side of the screen and upload an XTTS model that you want to finetune. If you are manually uploading a model, you would place your model
\n", - "#@markdown files in `models/xtts/{yourmodelfolderhere}` and you will need all the models files in that folder `config.json, dvae.pth, mel_stats.pth,`
\n", - "#@markdown `model.pth, speakers_xtts.pth, vocab.json`. Without 1x model available, Finetuning will not start.

\n", - "#@markdown Likewise you can download your finetuned model from there, OR copy it to your Google Drive after finetuning, for later use in AllTalk.

\n", - "#@markdown To access the Finetuning Gradio interface, connect to the `Google Colab Finetuning url` when Finetuning has started.
\n", + "#@markdown Click the `Play button` to the left of this message to start Finetuning
\n", "import re\n", "import time\n", "import json\n", @@ -389,37 +303,11 @@ "!python /content/alltalk_tts/finetune.py\n" ], "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, "cellView": "form", - "id": "Dewvq5s38Sfd", - "outputId": "30689b4e-2af2-4595-f2be-8bf42753f355" + "id": "Dewvq5s38Sfd" }, - "execution_count": 7, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Google Colab Finetuning url: https://assistance-amazing-periods-payday.trycloudflare.com\n", - "\n", - "********************************************************************\n", - "**** Use the above URL to connect to Finetuning on Google Colab ****\n", - "********************************************************************\n", - "\n", - "[FINETUNE] HTTP Request: GET https://checkip.amazonaws.com/ \"HTTP/1.1 200 \"\n", - "[FINETUNE] HTTP Request: GET https://api.gradio.app/pkg-version \"HTTP/1.1 200 OK\"\n", - "Running on local URL: http://127.0.0.1:7052\n", - "[FINETUNE] HTTP Request: GET http://127.0.0.1:7052/startup-events \"HTTP/1.1 200 OK\"\n", - "[FINETUNE] HTTP Request: HEAD http://127.0.0.1:7052/ \"HTTP/1.1 200 OK\"\n", - "\n", - "To create a public link, set `share=True` in `launch()`.\n", - "[FINETUNE] Received interrupt signal. Cleaning up and exiting...\n", - "\u001b[0m" - ] - } - ] + "execution_count": null, + "outputs": [] } ], "metadata": {