Spaces:
Build error
Apply for community grant: Personal project
This project connects Whisper to a speaker embedding model to add speaker recognition to Whisper. This makes Whisper much more usable to produce normal transcripts. People have reached out to me saying that the script I wrote with this program has helped them transcribe their new podcast or create a written document of their conversations with their grandparents. If I got a grant for a huggingface space gpu, it would allow so many more people to access this program via a simple drag-and-drop feature.
Here is the tweet where I announced the script which this space is based on:
https://twitter.com/dwarkesh_sp/status/1579672641887408129
And here is the tweet where I announced the creation of this space:
https://twitter.com/dwarkesh_sp/status/1601656163182948352
