-
Notifications
You must be signed in to change notification settings - Fork 302
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.
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
Opt-in consent #550
Comments
Diarization is so hit or miss I don't believe this is feasible right now. I was having a conversation with myself and it thought 4 speaker were talking. I've looking into diarization extensively for meeting minutes as well and nothing works even close to what you could call 'reliably'. Legal issues will be tough here and I think each user will have to make their own call - likely lots of us operating illegally for the foreseeable future. Not sure if a 'hands up' emoji is appropriate but that's where I believe we are today. |
I believe ethically we should have at least the ability for a consent mechanism (this might not technically easy at this point but I think we should be thinking about a mechanism).
We would have to identify known speakers.
Have them state their consent and optimal timeout for that consent.
In some countries there may be legal issues with recording I think giving the user the option to use this is a good move.
I've got eyes on Speechmatics flow it like a great technology for recognition and diarization.
The text was updated successfully, but these errors were encountered: