Generate audiobooks from E-books with Kokoro-82M

Re:Generate audiobooks from E-books with Kokoro-82M by swores on Wednesday January 15, 2025 @05:35AM (#42709430)

Can anyone recommend an open source option that would allow training on a custom voice (my own, so I'd be able to record as many snippets as it needed to train on) to allow me to use it for TTS generation without sharing it off my machine?

Edit: I'll wait to see if any recommendations get made here, if not I might give this one a go: https://github.com/coqui-ai/TTS

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by phrotoma on Wednesday January 15, 2025 @07:22AM (#42710099)

https://github.com/DrewThomasson/ebook2audiobook

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by numpad0 on Wednesday January 15, 2025 @07:07AM (#42709989)

I think you can probably generate TTS audio by classical means, and voice2voice that audio through RVC or Beatrice V2. Haven't looked into it in a while but Beatrice is apparently super fast and CPU only.

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by esskay on Wednesday January 15, 2025 @09:53AM (#42711444)

If I recall Coqui is very much a dead project, just one to be aware of.

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by hm64 on Wednesday January 15, 2025 @11:15AM (#42712786)

Coqui is great, but in practice, I found Piper easier to set up, train, and deploy as an ONNX file. Big thanks to the Sherpa development team for their helpful resources: https://k2-fsa.github.io/sherpa/onnx/tts/piper.html and to the Rhasspy team for their training guide: https://github.com/rhasspy/piper/blob/master/TRAINING.md.

I also found DEMUCS + Whisper + pydub to be a super helpful combo for creating quality datasets.

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by drewbitt on Wednesday January 15, 2025 @11:24AM (#42712945)

There is a fork here https://github.com/idiap/coqui-ai-TTS 'coqui-tts'

Though according to the TTS leaderboard, Fish Speech https://github.com/fishaudio/fish-speech and Kokoro are higher.

https://huggingface.co/hexgrad/Kokoro-82M

https://huggingface.co/fishaudio/fish-speech-1.5

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by xnx on Wednesday January 15, 2025 @02:27PM (#42715759)

AFAIK Kokoro can't be fine tuned

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by jsemrau on Wednesday January 15, 2025 @12:52PM (#42714433)

I wrote this a while ago about xTTSv2 mixed with Nvidia's Nemo. Maybe it kicks off your journey.

https://jdsemrau.substack.com/p/teaching-your-agent-to-speak...

[ Parent | on HN ]

Re:Generate audiobooks from E-books with Kokoro-82M by eamag on Monday January 20, 2025 @02:17PM (#42772026)

F5-TTS, I wrote a post about it https://eamag.me/2025/Voice-Cloning

[ Parent | on HN ]