I need more options for the model

[ehjr5u]

Describe the feature

Sometimes I need to access more configuration with the Whisper models because some options don't even exist with the basic configurations, and I don't to touch into the CLI to find my option

[thewh1teagle]

Vibe provide option to change every parameter of whisper. which options are missing?

[ehjr5u]

i'm looking for compression_ratio_threshold since i'm trying to fix the the repetitions of the same word that the models produce (like for example, one line contains "Hi", then the following lines repeat the same "Hi" when there's no talking

[ehjr5u]

and also "condition_on_previous_text" and "no_speech_threshold"

[ehjr5u]

Hello? are you gonna respond?