alt
Hacker News
spwa4
•
yesterday at 9:14 PM
•
0 replies
•
view on HN
These are native speech-to-speech models, so I think they've decided not to do that anymore.