Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by kenarsa on Tuesday August 05, 2025 @10:35PM
(#44807003)
Try https://github.com/Picovoice/orca It's about 7MB all included
[ Parent
| on HN ]
Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by gary_0 on Tuesday August 05, 2025 @11:04PM
(#44807147)
Not open source. "You will need internet connectivity to validate your AccessKey with Picovoice license servers ... If you wish to increase your limits, you can purchase a subscription plan." https://github.com/Picovoice/orca#accesskey
Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by cakealert on Wednesday August 06, 2025 @02:20AM
(#44808357)
Going online is a dealbreaker but if you really need it you could use ghidra to fix that. I had tried to find a conversion of their model to onnx (making their proprietary pipeline useless) but failed.
Hopefully open source will render them irrelevant in the future.
Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by papichulo2023 on Wednesday August 06, 2025 @02:57AM
(#44808588)
The guy is just spamming the project in a lot of comments.
Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by satvikpendem on Tuesday August 05, 2025 @10:38PM
(#44807026)
Does an apk for Android exist for replacing its speech to text engine? I tried sherpa-onnx but it was too slow for real time usage it seemed, and especially so for audiobooks when sped up.
Re:Kitten TTS: 25MB CPU-Only, Open-Source Voice Model
by satvikpendem on Tuesday August 05, 2025 @11:19PM
(#44807240)
I can't test this out right now, is this just a demo or is it actually an apk for replacing the engine? Because those are two different things, the latter can be used any time you want to read something aloud on the page for example. This is the sherpa-onnx one I'm talking about.