BoltzmannEntropy / xtts2-ui

A User Interface for XTTS-2 Text-Based Voice Cloning using only 10 seconds of speech
MIT License
242 stars 40 forks source link

Patch 1 #12

Closed lunar-studio closed 8 months ago

lunar-studio commented 8 months ago

Updated App2.py to include a basic ability to record from a microphone, changed a minor typo, updated the requirements to include the voice recording streamlit, and added a bat to help Windows users quickly launch the environment and app.py file.

To do (unfortunately, my time is very limited at present so anyone please feel free to help):

  1. I would like to also figure out a way to start app2.py from a bat file as well, but can only figure out a way to launch invoke it by calling streamlit which is installed by default into my local user's directory (Windows.) So off the top of my head it still needs to be manually started by the end user.
  2. I haven't edited the documention to reflect the added recording capability. I also figured you might want to reposition the text or location of the recording streamlit. Perhaps there's also a way to name it directly within the window and automatically populate the loading of a recording.