search

nextcloud / spreed

🗨️ Nextcloud Talk – chat, video & audio calls for Nextcloud
https://nextcloud.com/talk
GNU Affero General Public License v3.0
1.6k stars 428 forks source link

Transcription of live video calls #8503

Open sunjam opened 1 year ago

sunjam commented 1 year ago

Describe the solution you'd like Optional live closed caption transcription of a live video call. Allows for deaf and hard-of-hearing participation as they can read your words as you talk.

Video and Audio call transcriptions for the deaf and hard-of-hearing. Only platform I am using with live transcription support is Zoom.

Transcription support in this case is applied to the live conversation (as opposed to a recording).

Describe alternatives you've considered

Additional context I realize this might well be outside of the realm of what can be built into Talk, but I at least would love to collaborate on ideas and potential fixes. Thanks for considering either way! See this re-post to the forum

nickvergessen commented 1 year ago

The toughest issue will be to find a service that does transcription for free and has an acceptable outcome. Privacy respect being the follow up necessity.

nickvergessen commented 1 year ago

Not sure how feasible but whisper.cpp seems to be able to be compiled to WebAssembly: https://github.com/ggerganov/whisper.cpp/tree/master/examples/whisper.wasm

Originally posted by @juliushaertl in https://github.com/nextcloud/spreed/issues/3969#issuecomment-1368689728

Kofangun commented 5 months ago

is there news about that subject ? anyone ? sounds not too difficult with whisper