Bark TTS, Seamless Translation, RVC, Music Generation and More with the TTS Generation WebUI

[@nunuarthas8680]

As someone who's been painstaking adding inflection to TTS generations, Bark sounds really natural!

[@shawnvines2514]

Awesome video. Thank you very much for the work. Extremely informative and trustworthy content in a land of mostly algorithmic click bait or here's the same demo for the 10th time.

[@EthanJamesCarney]

Damn! I'm calling up my friend George Jetson to come over and show me how to use this. LOL

[@thegtlab]

Any idea how SUNO is able to produce such good sounding vocals for their songs? I wonder what TTS system they are using. Is it TTS into RVC?

[@grrinc]

Could anyone explain how you actually start the webui after it has loaded? I cannot see anything to click or see anything instructions in the video. Ta.

[@nanonomad]

If you used the one click installer, there should be a file in the base directory called start_windows.bat that will launch the WebUI in the proper virtual environment

[@grrinc]

@@nanonomad Thanks nano, but isn’t that the installer itself? Wouldn’t there be a separate ‘run’?

[@nanonomad]

Same script. It has checks to install miniconda and create the venv if it doesn't exist yet, then start the venv and launch the webui. The launch command it calls is python webui.py, but you'd need to do that from the correct venv if you weren't using the launch script

[@grrinc]

@@nanonomad Ahh okay, well it looks like it’s not working for me. I only get an ‘Done! Press any key to continue…’ message and then it all just shuts off. Thanks anyway, the hunt continues!

[@nanonomad]

If you haven't tried launching from a console window rather than double clicking it, try that.

[@alexsanders8881]

Is there any program we can use locally to translate downloaded videos?

[@nanonomad]

Whisper by openai is a ml model that does speech to text transcriptions and internal translation to other languages if you enable the option. Translation isnt supported by all the forks, but I think ones like faster-whisper do translation as well. Edit: I'm not sure what frontends are out there, but I'm sure there are some. It's pretty easy to use from the command like or through the python api.

[@christopherbrown1187]

no it does not. it fucking doesnt even installs everything withoutr an error. nothing works

[@yulang1221]

same