Jargonic Sets New SOTA for Japanese ASR

#1

Post by four_fifths » Wed, May 07, 2025, 12:21 PM UTC

Re: Jargonic Sets New SOTA for Japanese ASR

#2

Post by rfv6723 » Wed, May 07, 2025, 1:46 PM UTC

Why no comparition to gpt-4o-transcribe？

If you don't compare to latest model on the market, how can you claim it's SOTA?

According to OpenAI, gpt-4o-transcribe has much better performance than whisper-large-v2.

#3

Post by 1317 » Wed, May 07, 2025, 1:59 PM UTC

SOTA: not used in the article but probably State Of The Art

ASR: Automatic Speech Recognition, speech-to-text

#4

Post by albertzeyer » Wed, May 07, 2025, 2:03 PM UTC

Are there any details on what they changed to improve over other existing models?

#5

Post by lenerdenator » Wed, May 07, 2025, 2:13 PM UTC

1317 wrote: post #3

SOTA: not used in the article but probably State Of The Art ASR: Automatic Speech Recognition, speech-to-text

And here I was, as a ham radio operator, excited to read something about Summits On The Air.

shuffles dejectedly back to shack