Context size or maximum length of the audio
#3
by
esab
- opened
Hi,
Congrats for this amazing open audio model!
What's the maximum length of an input audio of the model? I've tested on the demo for a transcription task and the model only produces the first 30 seconds.
Yes, this version still supports audio input of up to 30 seconds only. Audio longer than 30 seconds will be truncated. In the near future, we will support longer audio.
what about the text length?128k?