erew123
commented
2 months ago Hi @Kafkanistan
I believe its because your folder name has a -main on the end of it.
D:\AI\text-generation-webui-main\extensions\alltalk_tts-main
and if you notice, its looking for the output in:
extensions/alltalk_tts/outputs/Amaya_1715268437.wav
Its an unfortunate annoyance of downloading off Github as a source code zip file (which is why I recommend a git clone as the best way to setup from Github). Python/Conda has some issues with dashes and spaces which cause these complications within the code handling.
If you rename your folder:
D:\AI\text-generation-webui-main\extensions\ alltalk_tts
and change your extension loading name in Text-gen-webui from alltalk_tts-main to alltalk_tts then all should be working for you.
Thanks
Warning Audio generation failed. Status: Error opening 'extensions/alltalk_tts/outputs/Amaya_1715268437.wav': System error.
Diagnotics log: diagnostics.log
Describe the bug It doesn't output any sound or generate or wav file. Warning Audio generation failed. Status: Error opening 'extensions/alltalk_tts/outputs/Amaya_1715268437.wav': System error.
To Reproduce Start Text generation webui, load in a model, load in a character. Go to chat and type.
Text/logs cmd.log
Desktop (please complete the following information): AllTalk was updated: [2024-05-09] Custom Python environment: [yes/no give details if yes]
Text-generation-webUI was updated: [[2024-05-09]
Additional context When I for example enable deespeed it says it loud DEEPSPEED ENABLED. Using the preview function does only generate the text in the cmd window but doesn't make any sounds.