//
sign in
Post
by @danabra.mov
PostEmbed
by @danabra.mov
Record
by @jimpick.com
Record
by @atsui.org
+ new component
Post
You have to choose the model. This was a crappy videocall, with somewhat noisy background and scientists speaking our version of English. ggml-medium.en-q8_0.bin seems to be the most accurate but still reasonably "small" model. You can download it from Hugging Face huggingface.co/ggerganov/wh... 3/
6d
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
ggerganov/whisper.cpp at main
Juan José García-Ripoll