Closed dhruvyad closed 2 weeks ago
Are you using response_format=text
? I am looking into this, and I believe you can work around this issue by using json
or verbose_json
response format types instead of text
.
This was fixed in the latest release (v0.9.0).
This line is throwing a TypeError:
when used for Whisper transcription. It can be fixed if you check that origin is a class: