Pre-processing with a vocal extraction model (bs-rofomer or similar) helps a lot... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		neckro23 3 months ago \| parent \| context \| favorite \| on: FFmpeg 8.0 adds Whisper support Pre-processing with a vocal extraction model (bs-rofomer or similar) helps a lot with the hallucinations, especially with poor quality sources.

trenchpilgrim 3 months ago [–]

I'm working with fairly "clean" audio (voice only) and still see ridiculous hallucinations.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact