←back to thread

High-fidelity simultaneous speech-to-speech translation

(arxiv.org)
112 points Bluestein | 3 comments | 03 Jul 25 20:27 UTC | HN request time: 0.604s | source
Show context
benlivengood ◴[03 Jul 25 21:09 UTC] No.44459218[source]▶
Now to get the model to run in an earbud...
replies(1): >>44459279 #
1. lapink ◴[03 Jul 25 21:18 UTC] No.44459279[source]▶
The model can actually run on an iPhone 16 Pro, so if the earbud is connected to one that could work!
replies(2): >>44459510 #>>44459922 #
2. Bluestein ◴[03 Jul 25 21:57 UTC] No.44459510[source]▶
That would be insane.-

Thinking of it, the whole "stack" from earbuds to phone to cloud - even in just something so "commonplace" as Assistant or Alexa ...

... Is amazing: All that computing power at our disposal.-

3. gcanyon ◴[03 Jul 25 23:21 UTC] No.44459922[source]▶
Almost as good as a babel fish!