Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

if i understood the wwdc keynote correctly, they completely redid the text-to-speech, so this should actually be good now (maybe)


Yeah, but they’re so far behind even if they 10x Siri they’ve got nothing on OpenAI.


Then again.. Siri is all on device. I don’t want to send my voice anywhere


The current voice transcription engine at OpenAI is using Whisper-1[0] which is open source and runnable locally, if you wanted to keep it all on-device. I run it locally for various things and it works pretty damn well.

[0] https://github.com/openai/whisper


I have used whisper but it seems very picky about which device it will perform well on ("well" meaning quickly)


So is Google’s version and it’s a 100x difference, but agree re OpenAI.


> Siri is all on device.

No it’s not. Try doing literally anything without an active internet connection and she’ll say “I’m having trouble connecting”




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

Search: