search

erew123 / alltalk_tts

AllTalk is based on the Coqui TTS engine, similar to the Coqui_tts extension for Text generation webUI, however supports a variety of advanced features, such as a settings page, low VRAM support, DeepSpeed, narrator, model finetuning, custom models, wav file maintenance. It can also be used with 3rd Party software via JSON calls.
GNU Affero General Public License v3.0
944 stars 110 forks source link

[QUESTION] How to run script.py with Custom Command Line Arguments? #221

Closed deffcolony closed 4 months ago

deffcolony commented 4 months ago

I'm currently running the following commands to execute alltalk_tts:

conda activate alltalk
python script.py

I need to know if there's support for running the command with custom command line arguments, similar to what XTTS offers, as demonstrated below:

python -m xtts_api_server --device cuda -hs 0.0.0.0 --deepspeed --use-cache --listen

Would appreciate some guidance on how to use custom command line arguments into the execution of script.py with alltalk_tts.

erew123 commented 4 months ago

Hi @deffcolony

There are currently no command line arguments at the moment. Typically if you are running AllTalk as a standalone, you would run AllTalk's setup script which will create a custom python environment and then batch or shell script files to start the Python environment and start AllTalk e.g.

image

The IP address cannot be fixed to 0.0.0.0 at the moment, though you can manually set it to an IP address: in the web page interface http://127.0.0.1:7851/ where you can also set the start-up behaviour of DeepSpeed etc. Beyond that, if you have not activated them, you can call for them using the API calls https://github.com/erew123/alltalk_tts?tab=readme-ov-file#-api-suite-and-json-curl

image

Things such as binding to 0.0.0.0 are being addressed in version 2 of AllTalk. https://github.com/erew123/alltalk_tts/discussions/211

Thanks