Repolode
Submit
idiap

idiap/coqui-ai-TTS

Open-source Python toolkit for text-to-speech and voice conversion, with pretrained models, training tools and dataset curation utilities.

Open-source alternative to:ElevenLabsGoogle Cloud Text-to-Speech

View on GitHubWebsite
Stars
2,337
Forks
298
Open issues
17
Last commit
2 days ago

What is it?

Coqui TTS is a deep learning library for generating speech from text. It ships pretrained models covering more than 1100 languages, and it provides code to train new models or fine-tune existing ones on your own data. A fork of the original Coqui TTS project, it is maintained under the coqui-tts PyPI package.

The library implements a range of architectures, including Tacotron2, Glow-TTS, FastPitch, VITS, and the XTTS model, which supports voice cloning and streaming output. It also includes vocoders such as HiFiGAN and WaveRNN, and voice conversion models like FreeVC, kNN-VC and OpenVoice. Multi-speaker and multilingual synthesis are supported.

Users can call the library through a command-line tool named tts, a Python API, or a Docker image that runs a local TTS server. A dataset_analysis directory offers tools for inspecting and curating speech datasets.

Key features

  • 01Pretrained text-to-speech models for more than 1100 languages.
  • 02Voice conversion models including FreeVC, kNN-VC and OpenVoice.
  • 03Training and fine-tuning recipes for new or existing TTS models.
  • 04Multi-speaker and multilingual synthesis support.
  • 05Command-line tool, Python API and Docker-based TTS server.
  • 06Dataset analysis and curation utilities.

Use cases

  • Adding speech output to an application in many languages using pretrained models.
  • Cloning a voice from a short reference audio clip with the XTTS model.
  • Training or fine-tuning a custom TTS voice on your own recorded dataset.
  • Running a local TTS server for internal tools without relying on a cloud API.

Quick start

  1. 1

    Install PyTorch, torchaudio and torchcodec (uv picks the right build for your system)

    uv pip install torch torchaudio torchcodec --torch-backend=auto
  2. 2

    Install coqui-tts from PyPI

    uv pip install coqui-tts
  3. 3

    List the available pretrained models from the command line

    tts --list_models

Things to know

  • PyTorch is not included by default since coqui-tts 0.27.4, so you must install it yourself first.
  • Officially tested on Ubuntu 24.04 with Python 3.10 to 3.14; Mac and Windows should work but are less tested.

Topics

#text-to-speech#voice-conversion#deep-learning#pytorch#speech-synthesis#voice-cloning#python

FAQ

What is coqui-ai-TTS?
Open-source Python toolkit for text-to-speech and voice conversion, with pretrained models, training tools and dataset curation utilities. Coqui TTS is a deep learning library for generating speech from text. It ships pretrained models covering more than 1100 languages, and it provides code to train new models or fine-tune existing ones on your own data. A fork of the original Coqui TTS project, it is maintained under the coqui-tts PyPI package.
Is coqui-ai-TTS free to use? What license does it have?
coqui-ai-TTS is open source under the MPL-2.0 license. Check the license terms before using it in commercial products.
Is coqui-ai-TTS still maintained?
Actively maintained. The last commit to coqui-ai-TTS was on Oct 19, 2026.
What can I use coqui-ai-TTS for?
Adding speech output to an application in many languages using pretrained models. Cloning a voice from a short reference audio clip with the XTTS model. Training or fine-tuning a custom TTS voice on your own recorded dataset. Running a local TTS server for internal tools without relying on a cloud API.
How do I install coqui-ai-TTS?
Install PyTorch, torchaudio and torchcodec (uv picks the right build for your system): uv pip install torch torchaudio torchcodec --torch-backend=auto Install coqui-tts from PyPI: uv pip install coqui-tts List the available pretrained models from the command line: tts --list_models
Where can I get coqui-ai-TTS?
The source code and installation instructions are on GitHub: https://github.com/idiap/coqui-ai-TTS

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects