kromskii2/alltalk_tts
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
630 Commits 2 Branches 23 Tags
92ce5e24bc9d8d7341b705de87e0cb16cda033ff
Commit Graph
630 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
erew123 1c29a8391d Update README.md 
DeepSpeed 11.2 instructions
 2023-12-14 06:42:03 +00:00
erew123 d9cf355b79 Update README.md 2023-12-13 12:28:04 +00:00
erew123 3762ccd396 Update README.md 2023-12-13 12:19:44 +00:00
erew123 7c8e57d6b3 Update README.md 2023-12-13 12:15:03 +00:00
erew123 e8dde29d20 Documentation update 
Updated a note in the documentation about correct formatting of RP character greeting cards
1.4 		2023-12-13 12:11:28 +00:00
erew123 8d7b7018ae Update README.md 2023-12-13 11:51:12 +00:00
erew123 2262046fd0 Update README.md 2023-12-13 11:50:28 +00:00
erew123 bdc74d3333 Update README.md 2023-12-13 11:42:46 +00:00
erew123 8c8e65fb1d Fixed outlier on character paragraph. 
On new chat, character paragraphs, found an outlier possibility where character/narrator wouldn't swap, so performed replace('&quot;<. *', '&quot;< *"').replace('< *"', '< *') which should capture it.
 2023-12-13 10:00:33 +00:00
erew123 fedea7cc09 Update README.md 2023-12-13 06:52:27 +00:00
erew123 992e9625ef Update README.md 2023-12-13 05:48:58 +00:00
erew123 36f44fce11 Update README.md 2023-12-13 05:39:30 +00:00
erew123 e12cf5f0a2 Update README.md 2023-12-13 05:28:46 +00:00
erew123 b7de2988b0 Updated note about clip length 
Turns out the models default config file is setting "max_ref_len": 30, allowing for up to 30 second long voice clips to be used.
 2023-12-13 05:21:38 +00:00
erew123 bbfbeba246 Update README.md 2023-12-13 04:58:01 +00:00
erew123 40b775459e New Narrator Version 
The narrator and filtering has been re-written to vastly improve filtering of text, resulting in less unwanted sounds and better control over narrator/character.
 2023-12-13 04:51:41 +00:00
erew123 53b314d127 Update requirements_other.txt 
lower version of numpty+python therefore
v1-old-narrator 		2023-12-12 23:56:29 +00:00
erew123 695bcca7c8 Update requirements_nvidia.txt 
allow lower version of python
 2023-12-12 23:56:00 +00:00
erew123 d35d189d11 Update config.json 
increased the repetition penalty from 10 to 12 as its keeping the voice more in character. (possible 11 may be ok too)
 2023-12-11 20:41:07 +00:00
erew123 fc3bb6f406 Update tts_server.py 
small spelling mistake
 2023-12-11 19:22:41 +00:00
erew123 a6cfe15c94 Update README.md 2023-12-11 13:51:19 +00:00
erew123 e4b581a1be Backend access to temp and repetition 
Added the capability to change the default temperature and repetition settings of the model along with writing documentation to explain it all.
 2023-12-11 13:44:12 +00:00
erew123 f42b2c1d26 A few more voices. 
Creative commons 3 from https://commons.wikimedia.org/ hence original sample details included
 2023-12-11 08:09:40 +00:00
erew123 61dfb5402f Update README.md 2023-12-11 08:06:24 +00:00
erew123 567030079f Update README.md 2023-12-11 08:03:41 +00:00
erew123 00178c0188 added some more voices 
Creative commons 0
 2023-12-11 06:55:53 +00:00
erew123 803c58b44e Extra tweak to the text preprocessing 
Discards any text to be generated that is 0 or 1 characters in length after the sentence is split down (usually this is a * or a " and just produces a strange sound).
 2023-12-11 04:15:52 +00:00
erew123 07b32d0190 Update README.md 2023-12-11 03:46:59 +00:00
erew123 ee3f5cb853 Applied extra settings to TTS generation 
Pulled some extra settings from the models configuration file to help keep generated speech closer to the sample wav.
 2023-12-11 02:53:51 +00:00
erew123 6ea56eddfa Allow standalone running 
Added logic to allow script to be run as a standalone if required (outside of Text-generation-webui)
Corrected spelling mistake in documentation.
 2023-12-10 20:02:36 +00:00
erew123 963f06cc7f Update README.md 2023-12-10 11:04:38 +00:00
erew123 bc9a5d7939 Update README.md 2023-12-10 10:59:39 +00:00
erew123 6b00565179 Delete requirements.txt 
added new requirements files
 2023-12-10 10:55:46 +00:00
erew123 064f4e34c7 Update README.md 2023-12-10 10:54:55 +00:00
erew123 aa889bc897 Update README.md 2023-12-10 10:53:14 +00:00
erew123 5124c6c6b3 Test around cuda for non-cuda users 
- made logic for model loading on non-cuda systems
- split requirements files into nvidia and other
- updated other files as necessary
 2023-12-10 10:48:48 +00:00
erew123 9927b51622 Populate settings with wav voices 2023-12-10 09:49:07 +00:00
erew123 3264dfecde Voices list now populated from the folder 2023-12-10 09:48:10 +00:00
erew123 d5836bffd6 Add files via upload 2023-12-10 09:04:51 +00:00
erew123 ff4ae90eb5 Adding a few more voices 
Added a couple more voices from Freesound on a Creative Commons 0 license https://creativecommons.org/publicdomain/zero/1.0/
 2023-12-10 09:04:11 +00:00
erew123 579115885a Narrator more robust 
Many of the LLM's aren't using asterisks to correctly split the narration from the spoken text. Have re-written the entire text pre-processing engine to take text within asterisks as being narrated text, text within quotes as the characters voice and to assume that anything else is narrated text (the bits that the LLM's aren't correctly using asterisks with). Also added some additional filtering to clear up some strange text characters that may be sent to the TTS engine.
 2023-12-10 07:58:44 +00:00
erew123 c5c1a455df Minor update to deal with gr.dropdown 
resolves complaints about custom messages in the dropdowns for language, default & narrator voice.
 2023-12-10 01:57:00 +00:00
erew123 c681cd7e4d Update README.md 2023-12-10 00:29:47 +00:00
erew123 ff6b4be5b5 Update README.md 2023-12-10 00:29:23 +00:00
erew123 9b9f65b78d Update README.md 2023-12-09 21:21:18 +00:00
erew123 e6445ea974 Update README.md 2023-12-09 21:20:37 +00:00
erew123 24b1a6333a Update README.md 2023-12-09 21:18:05 +00:00
erew123 df424cead4 Update README.md 2023-12-09 21:14:30 +00:00
erew123 d7677c0a89 Update README.md 2023-12-09 21:12:49 +00:00
erew123 6d7a96dc68 Update README.md 2023-12-09 21:08:13 +00:00