- cross-posted to:
- framasoft@lemmy.world
- cross-posted to:
- framasoft@lemmy.world
You must log in or # to comment.
I’ll wait for a local model where the transcription happens on-device, like Google Recorder.
I like their donation model. I hope they success.
This is about a speech transcription program running on the company’s remote server. The app uploads your audio and the system sends you a transcript. You’ve got to be kidding. Reporting as spam.
It’s free software which you can host yourself. The source is here (GPLv3). You can read more about the people that make it here: https://en.wikipedia.org/wiki/Framasoft
It didn’t say that on the linked page. Is the AI model and training code and data also free?
Added: it looks like it uses Whisper for transcription. So the inference code is there but it’s unclear about the other stuff.
https://en.m.wikipedia.org/wiki/Whisper_(speech_recognition_system)
Anyway, thanks for the update.
Framasoft
They’re also involved in Fediverse development, they made Mobilizon as an event management platform
https://framablog.org/2022/11/08/mobilizon-v3-find-events-and-groups-throughout-the-fediverse/
