Purfview
Purfview
@SeBL4RD Looks like your described issue has nothing to do with the the topic.
> Oh? Could you enlighten me? I admit that a little help would be welcome. I think your issue is same as issue "1175".
>Oh, and I'm not using batch mode. Then it definitely has nothing to do with the topic. You should open a new issue.
Compare to Openai-whisper
In that table openai-whisper is similar to faster-whisper. Better ask Speechbrain what settings trigger different result.
Better ask Speechbrain, they will know more about it. Here the aim is more or less to replicate the openAI not some Speechbrain's function.
@hforghani >I just have used BatchedInferencePipeline and it got solved: Then I think this solved your problem: `condition_on_previous_text=False` @Emre-Akgul >Got same problem for Armenian and Georgian languages. Try: `condition_on_previous_text=False` `word_timestamps=False`...
Could you share the audio sample to reproduce the issue?
purfview [@] protonmail [.] com
I didn't got any repeats, my settings used: `--device=cpu --language=de --model=large-v2 --compute_type=float32 --beam_size=5 --vad_filter=False` Make sure you are using the latest `0.7.1` version.