Re:Opus 1.5 released: Opus gets a machine learning upgrade
by indolering on Monday March 04, 2024 @11:01PM
(#39599273)
It's a really smart application of ML: helping around the edges and not letting the ML algo invent pheonems or even whole words by accident. ML transcription has a similar trade-off of performing better on some benchmarks but also hallucinating results.
[ Parent
| on HN ]
Re:Opus 1.5 released: Opus gets a machine learning upgrade
by kolinko on Tuesday March 05, 2024 @01:51AM
(#39600183)
A nice story about Xerox discovering this issue in 2003, when their copiers began slightly changing random numbers in copied documents
https://www.theverge.com/2013/8/6/4594482/xerox-copiers-rand...
Re:Opus 1.5 released: Opus gets a machine learning upgrade
by cedilla on Tuesday March 05, 2024 @03:05AM
(#39600587)
I don't think machine learning was involved there at all. As I understand it, it was an issue of a specifically implemented feature (reusing a single picture of a glyph for all other instances to save space) being turned on in archive-grade settings, despite the manual stating otherwise.
Re:Opus 1.5 released: Opus gets a machine learning upgrade
by h4x0rr on Tuesday March 05, 2024 @02:51AM
(#39600504)
https://youtu.be/zXXmhxbQ-hk
Interesting yet funny CCC video about this
Re:Opus 1.5 released: Opus gets a machine learning upgrade
by yalok on Tuesday March 05, 2024 @03:26AM
(#39600721)
fwiw, in ASR/speech transcription world, it looks reverse to me - in the past, there was lots of custom non-ML code & separate ML models for audio modeling and language modeling - but current SOTA ASRs are all e2e, and that's what's used even in mobile applications, iiuc.
I still think the pendulum will swing back there again, to have even better battery/larger models on mobile.