News for hackers. Stuff that matters.
Re:OpenAI Audio Models by claiir on Saturday March 22, 2025 @06:12AM (#43444690)
(2) is also significantly harder with these new models as they don’t support word timestamps like WHISPR.

see > Other parameters, such as timestamp_granularities, require verbose_json output and are therefore only available when using whisper-1.

There are no comments on this comment yet.