音频与语音MPL-2.0Python

Coqui TTS

Deep-learning text-to-speech toolkit, including the XTTS voice cloning model.

Star 数
46.1K
Fork 数
6.2K
未关闭 Issue
3
最近提交
2024-08-16

数据来自 GitHub API · 更新于 2026-10-15

Coqui TTS 是什么?

Coqui TTS provides many pretrained TTS models and training recipes, including XTTS-v2 for multilingual voice cloning from a few seconds of audio. Coqui the company has shut down; the community fork published as coqui-tts on PyPI continues maintenance.

主要功能

  • Dozens of pretrained voices and languages
  • XTTS-v2 voice cloning from a short sample
  • Training and fine-tuning recipes
  • CLI and local server

系统要求

平台
Windows · macOS · Linux
显卡
Optional; NVIDIA recommended for XTTS
Python
3.10–3.12 (fork)

如何安装 Coqui TTS

根据项目 README 整理。开源项目更新很快,请以 GitHub 页面上的最新说明为准。

pip (maintained fork)

  1. 1
    Install the community-maintained package.
    pip install coqui-tts
  2. 2
    Synthesize speech.
    tts --text "Hello from open-source TTS" --out_path speech.wav

优势

  • Huge model zoo
  • Voice cloning

不足

  • Original repo no longer maintained
  • XTTS weights use the non-commercial Coqui Public Model License

Coqui TTS 替代品对比

软件Star 数许可证平台显存
Coqui TTS

Deep-learning text-to-speech toolkit, including the XTTS voice cloning model.

46.1KMPL-2.0—
F5-TTS

Zero-shot voice cloning TTS based on flow matching.

15.4KMIT—
Kokoro

A tiny 82M-parameter TTS model with surprisingly natural voices.

9.3KApache-2.0—
GPT-SoVITS

Few-shot voice cloning and TTS with a full training WebUI.

62.8KMIT6 GB+
Fish Speech

Multilingual TTS and voice cloning from Fish Audio.

33.1KApache-2.04 GB+
Whisper

OpenAI’s open-source speech recognition and translation model.

110.5KMIT~1 GB (tiny) to ~10 GB (large)

常见问题

Coqui TTS 免费吗?

是的——Coqui TTS 是开源软件,采用 MPL-2.0 许可证。商用前请确认软件及模型权重的许可条款。

Coqui TTS 支持 Windows、Mac 和 Linux 吗?

Coqui TTS 支持:Windows, macOS, Linux。

运行 Coqui TTS 需要什么配置?

Optional; NVIDIA recommended for XTTS

如何安装 Coqui TTS?

1. Install the community-maintained package. 2. Synthesize speech.

Coqui TTS 有哪些好的替代品?

常见替代品包括 F5-TTS, Kokoro, GPT-SoVITS, Fish Speech,详见本页对比表。