←back to thread

High-fidelity simultaneous speech-to-speech translation

(arxiv.org)
112 points Bluestein | 1 comments | 03 Jul 25 20:27 UTC | HN request time: 1.245s | source
Show context
Grosvenor ◴[03 Jul 25 22:25 UTC] No.44459656[source]▶
This is so cool. The future is cool!

I wonder how it will work on languages that have different grammatical structure than french/english? Like Finno-Ugric languages which have sort of a Yoda speech to them. Edit: In Finno-Ugric languages words later on in a sentence can completely change the meaning. Will be interesting to look at.

It's considerate of them to name it after my favourite whisky.

replies(3): >>44459733 #>>44460606 #>>44461787 #
1. lapink ◴[03 Jul 25 22:38 UTC] No.44459733[source]▶
The alignment between source and target is automatically inferred, basically by searching when the uncertainty over a given output word reduces the most once enough input words are seen. This is then lifted to the audio domain. In theory the same trick should work even with longer grammatical inversions between languages, although this will lead to larger delays. To be tested!