Moonshine, the new state of the art for speech to text

Re:Moonshine, the new state of the art for speech to text by spencerchubb on Sunday October 27, 2024 @10:41AM (#41962910)

So they're claiming SOTA because they compare against OpenAI as SOTA. What about Groq or fal.ai?

[ Parent | on HN ]

Re:Moonshine, the new state of the art for speech to text by wut42 on Monday October 28, 2024 @07:57AM (#41970179)

Groq and fal are hosters of third party models. They cannot be SOTA.

Fal provides Whisper which is OpenAI.

Groq only hosts LLMs.

[ Parent | on HN ]

Re:Moonshine, the new state of the art for speech to text by rjwilmsi on Monday October 28, 2024 @10:10AM (#41971173)

My take wasn't that they are claiming SOTA, just that they are claiming same/better accuracy than whisper "tiny" and "base" models at higher performance versus original whisper.

The WER scores on the larger whisper models are lower (better) than "tiny"/"base", so none of these small models are close to SOTA either way.

[ Parent | on HN ]

Re:Moonshine, the new state of the art for speech to text by jibinmathew on Thursday October 31, 2024 @02:21AM (#42003984)

Groq and Fal.ai just companies that host, they dont make models

[ Parent | on HN ]