News for hackers. Stuff that matters.

A comment on Show HN: WhisperFusion – Ultra-low latency conversations with an AI chatbot. Parent comment

Re:Show HN: WhisperFusion – Ultra-low latency conversations ... by dmw_ng on Monday January 29, 2024 @03:06PM (#39181736)
The latency on this (or lack thereof) is the best I've seen, would love to know more about how it's achieved. I asked the bot and it claimed you're using Google's speech recognition, which I know supports streaming, but this result seems much lower lag than I remember Google's stuff being capable of
[ Parent | on HN ]
Re:Show HN: WhisperFusion – Ultra-low latency conversations ... by selcuka on Monday January 29, 2024 @06:35PM (#39184270)
> I asked the bot and it claimed you're using Google's speech recognition

That doesn't sound plausible. How can the LLM part know which speech recognition service is being used?

[ Parent | on HN ]
Re:Show HN: WhisperFusion – Ultra-low latency conversations ... by meandmycode on Monday January 29, 2024 @06:59PM (#39184486)
It's not entirely unlikely that the llm is informed exactly what it's source data is, with the hope that it can potentially make corrections to transcription errors
[ Parent | on HN ]
Re:Show HN: WhisperFusion – Ultra-low latency conversations ... by lolinder on Monday January 29, 2024 @08:14PM (#39185158)
Or just because it's interesting and people might ask. I could imagine it being a hallucination, but it could also be an easter egg of sorts.
[ Parent | on HN ]
Re:Show HN: WhisperFusion – Ultra-low latency conversations ... by selcuka on Tuesday January 30, 2024 @01:38AM (#39187042)
Apparently it uses the Web Speech API [1], not a specific service.

[1] https://developer.mozilla.org/en-US/docs/Web/API/SpeechRecog...

[ Parent | on HN ]