Whisper API only transcribing first few seconds

[_j]

@awissink

this is definitely an API-related bug

Every other platform works fine except the Apple stack. Are you sure you don’t want to blame Apple and their audio encoding?

Here’s that link to a link:

Whisper API is not able to transcribe audios created on iOS

You actually have failing audio files logged for analysis and they are understandable but can’t be transcribed? Here I describe a re-encoding you could do, which also has the effect of recoding in voice-over-ip audio bandwidth, so if there was something like noise shaping in high definition audio, it would be stripped. https://community.openai.com/t/sending-an-hours-worth-of-audio-through-whisper-using-node-js/450869/8