i got ouput like "[sounds of a bus]" and "[Side-by-side conversation]" "[Pause]" ,does anyone know why?thx

[tokyo4]

[SIDE CONVERSATION]
[Sounds of a man talking in the background]

Is it because of poor sound quality, such as low volume? Thank you

in the real time sound,this output also make i lose the ASR output text before.

[misutoneko]

Never seen those particular sound descriptions but I suppose poor quality can do that.
What was the actual command line? Because as of right now, it isn't even clear which model was used.
The smaller models are more likely to miss things, so switching to a bigger model may help.

If you can post a sample that might clear things up as well.