New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Whisper microphone example: do not repeatedly acquire and release the microphone #1864

Open

lavafroth wants to merge 2 commits into huggingface:main from lavafroth:acquire-microphone

lavafroth commented Mar 18, 2024

Changes

The input stream from the microphone is built once in the thread that handles it.
Inside the loop block, the stream is played and paused for the defined duration (300ms) to capture the pcm data.
The iter::repeat(()).enumerate() to keep track of the inference iteration is replaced with a simple 0.. range.

lavafroth added 2 commits

March 18, 2024 09:51


          acquire the microphone once to play and pause as needed

708dbee


          use a range instead of enumerating iter::repeat

1ea56ee

krzysztofwos mentioned this pull request

Whisper microphone example outputs gibberish #2182

Open

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment