←back to thread

High-fidelity simultaneous speech-to-speech translation

(arxiv.org)
112 points Bluestein | 2 comments | 03 Jul 25 20:27 UTC | HN request time: 0.433s | source
1. wedn3sday ◴[03 Jul 25 21:28 UTC] No.44459355[source]▶
For anyone else looking for examples: https://huggingface.co/spaces/kyutai/hibiki-samples
replies(1): >>44462923 #
2. littlestymaar ◴[04 Jul 25 09:38 UTC] No.44462923[source]▶
The high fidelity examples (see CFG-10 in the page) where the translated version has a very heavy French accent is kind of impressive (not that it is really useful, but impressive indeed).