search

mudler / LocalAI

:robot: The free, Open Source alternative to OpenAI, Claude and others. Self-hosted and local-first. Drop-in replacement for OpenAI, running on consumer-grade hardware. No GPU required. Runs gguf, transformers, diffusers and many more models architectures. Features: Generate Text, Audio, Video, Images, Voice Cloning, Distributed, P2P inference
https://localai.io
MIT License
25.84k stars 1.94k forks source link

[feature] TTS support request #1819

Open PylotLight opened 8 months ago

PylotLight commented 8 months ago

Is your feature request related to a problem? Please describe. Adding support for extra TTS models/backends?

Describe the solution you'd like Add support for these models: https://github.com/yl4579/StyleTTS2 https://github.com/myshell-ai/OpenVoice https://github.com/collabora/WhisperSpeech

Describe alternatives you've considered Perhaps I don't understand backend vs model so not sure if these are already supported or not.

Additional context N/A

localai-bot commented 7 months ago

Hello PylotLight,

Thank you for submitting your feature request for TTS support in LocalAI. I am a bot, an experiment of @mudler, and I will try to assist you with your issue.

From your request, it seems you would like to see support for additional Text-to-Speech (TTS) models in LocalAI. You have mentioned three models:

  1. StyleTTS2: https://github.com/yl4579/StyleTTS2
  2. OpenVoice: https://github.com/myshell-ai/OpenVoice
  3. WhisperSpeech: https://github.com/collabora/WhisperSpeech

To clarify, LocalAI is a versatile tool that supports various models, and it would be great to have these TTS models integrated in case they are not yet supported. If you are unsure whether a specific model is already supported, you can check the official LocalAI documentation or the models list in the LocalAI GitHub repository.

If you are interested in contributing to the LocalAI project and providing support for these TTS models, you can follow the contribution guidelines specified in the LocalAI repository.

In case you do not have the technical means to contribute, you can simply upvote this issue and tag @mudler or one of the other maintainers to review and possibly work on this enhancement.

Let me know if you have any more questions or need further assistance.