Hacker News new | past | comments | ask | show | jobs | submit login
Nvidia Speech and Translation AI Models Set Records for Speed and Accuracy (nvidia.com)
38 points by belter 14 days ago | hide | past | favorite | 3 comments



I've been using WhisperDesktop ( https://github.com/Const-me/Whisper ) with great success on a 3090 for fast & accurate transcription of often poor quality euro-english hours long multispeaker audio files. If there's an easy way to compare I'm certainly going to give this a try.

Relatively unknown , but Android text to speech in Google Gboard and Google assistant have gotten waaay better.

Cool to see that this beats Whisper while lowering latency. I'm happy TTS models are still "open by default" thus far.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: