Re:Ask HN: Real-time speech-to-speech translation
by billylo on Friday October 25, 2024 @10:14AM
(#41945526)
Author of 3PO here: check out our latest version 2.12. Many fixes have been incorporated in the past two weeks. Cheers.
[ Parent
| on HN ]
Re:Ask HN: Real-time speech-to-speech translation
by thangalin on Friday October 25, 2024 @01:56PM
(#41947754)
> Many fixes have been incorporated in the past two weeks.
Thanks for the efforts! Still many fixes to go, though: I used the version from two days ago, which had numerous issues. Also, 3PO isn't offline, so I won't be pursuing it.
Re:Ask HN: Real-time speech-to-speech translation
by billylo on Friday October 25, 2024 @05:56PM
(#41950210)
No problem! Some fixes were on the server side. We had a server side issue a couple days ago for a few hours. You may have been affected by it, giving you those errors. They have been fixed too. Take care!
Re:Ask HN: Real-time speech-to-speech translation
by supermatt on Tuesday October 29, 2024 @02:28AM
(#41980002)
It works really well for me, but I wish it supported more languages for the input - i guess this is a limitation of the model you are using? Do you mind giving some info about the tech stack? I love how fast it is.
Re:Ask HN: Real-time speech-to-speech translation
by billylo on Tuesday October 29, 2024 @03:46AM
(#41980477)
thanks for the kind words!! It is built on top of a fairly straightforward infrastructure. Client side is C#, .NET, MAUI. Server side is Firebase Cloud Functions/Firestore/Web Hosting + Azure.
Re:Ask HN: Real-time speech-to-speech translation
by supermatt on Tuesday October 29, 2024 @07:11AM
(#41982652)
Thanks - I was wondering more about the translation pipeline - i am assuming something like whisper and m2m100? How did you manage to get the latency so low? Other examples I have seen feel really sluggish in comparison.