ChatGPT's voice, self-hosted?

mrashley OP , 9 days ago

Oh WOW! Thanks to all who commented. Next time I get a chance I’m going to check these all out! 👍🏻 I hope others find this thread helpful too!

pythia , 10 days ago

i use these two all the time for tts:

github.com/JarodMica/ai-voice-cloning / github.com/gitmylo/audio-webui

B0rax , 10 days ago

You mean just the text to speech part? Look into Piper

PeachMan , 10 days ago

epub2tts: github.com/aedocw/epub2tts

Looks like a project that utilizes coqui-AI: github.com/coqui-ai/TTS

Ashyr , 10 days ago

When can I get one of these voices to read an epub on my phone? I’d love to have something like that

Lemongrab , 9 days ago

Librera FD as your reader app: www.f-droid.org/en/…/com.foobnix.pro.pdf.reader/
Sherpa Onnx as your TTS engine: github.com/k2-fsa/sherpa-onnx

I recommend the piper TTS pretrained models, either Lessac medium or Kusal high/medium

Lemongrab , 9 days ago

Installing Sherpa Onnx TTS makes it an option to use as your system TTS voice

kakes , 10 days ago

Regarding the TTS specifically, I remember looking into TorToiSeTTS back when this stuff was first coming out. You can generate ElevenLabs quality audio with it, but it’s insanely slow. In fact, when I was looking into it, it seemed like ElevenLabs may have been using a (much faster at the time) version of TorToiSe TTS, given the output is so similar.

According to the linked Github page, they seem to have solved the speed issues now, so it might be worth looking into. Of course, the other commenters have provided solutions that are pre-integrated into the LLM, but if you’re just looking for TTS this could be worth checking out. Also worth noting that this requires an NVIDIA GPU.

ReedReads , 10 days ago

Localai.io

github.com/mudler/LocalAI

requiem , 10 days ago

github.com/vndee/local-talking-llm

anzo , 10 days ago

This is what OP looks for. It exists! Other repos only cover partially (e.g. either ollama or tts)