Skip to content

I was adding stuff, but ended up rewriting the whole script#1

Open
Blasted246 wants to merge 11 commits intoLazycoder744:mainfrom
Blasted246:main
Open

I was adding stuff, but ended up rewriting the whole script#1
Blasted246 wants to merge 11 commits intoLazycoder744:mainfrom
Blasted246:main

Conversation

@Blasted246
Copy link

@Blasted246 Blasted246 commented Aug 29, 2025

Changes

Added

  • Won't repeat username for repeat messages until a certain amount of time
  • Romanization, for japanese usernames mostly
  • Exclude apps from attenuation
  • Queue for messages
  • Change the tts voice
  • Full japanese prononciation support
  • Different voice per language
  • Correct prononciation for Nya and Nani

Removed

  • Removed OAuth connect with just CHANNEL_NAME.
  • Removed startup scripts since they aren't needed.
  • Removed espeak-ng

Changed

  • Edge TTS only
  • Faster pipeline: two-stage async (generate → play) to reduce gaps between messages.
  • Attenuation smooth fade in/out; change duration with ATTENUATION_DELAY_MS.
  • Delete temp files after playback
  • requirements.txt updated
  • README.md updated to match current
  • Moved license to the correct file format for github
  • Added myself to license

Blasted246 and others added 11 commits August 28, 2025 22:52
I kinda just rewrote the script
this script is not offline, it reads twitch chat, which needs network
Better japanese support, using second tts model and swapping dynamically.
Filter emotes.
added exceptions for nya and nani
better JP support, and nya, nani is pronounced correctly.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant