音频与语音MITPython

Stable Audio Tools

Training and inference code for Stable Audio Open.

Star 数
3.9K
Fork 数
485
未关闭 Issue
131
最近提交
2026-10-09

数据来自 GitHub API · 更新于 2026-10-15

Stable Audio Tools 是什么?

Stable Audio Tools provides the code to run Stable Audio Open — a text-to-audio model for sound design, loops and short music — plus training utilities for conditional audio generation.

主要功能

  • Text-to-audio for SFX, loops and ambience
  • Gradio interface
  • Training and fine-tuning scripts

系统要求

平台
Linux · Windows
显卡
NVIDIA
Python
3.8+

如何安装 Stable Audio Tools

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

  1. 1
    Install the package.
    pip install stable-audio-tools
  2. 2
    Run the Gradio app from a clone of the repo with a pretrained model.
    python3 ./run_gradio.py --pretrained-name stabilityai/stable-audio-open-1.0

优势

  • Good for sound design

不足

  • Model license requires accepting Stability’s terms

Stable Audio Tools 替代品对比

软件Star 数许可证平台显存
Stable Audio Tools

Training and inference code for Stable Audio Open.

3.9KMIT—
AudioCraft (MusicGen)

Meta’s library for MusicGen music generation and AudioGen sound effects.

23.7KMIT16 GB for the large model; smaller models need less
Bark

Suno’s generative audio model: speech, music snippets and sound effects.

39.3KMIT~12 GB full; ~8 GB with small models
Whisper

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

110.5KMIT~1 GB (tiny) to ~10 GB (large)
whisper.cpp

Whisper in plain C/C++ — fast on CPUs, Apple Silicon and mobile.

54.4KMIT—
faster-whisper

Whisper reimplemented on CTranslate2 — up to 4× faster with less memory.

25.9KMIT—

常见问题

Stable Audio Tools 免费吗?

是的——Stable Audio Tools 是开源软件,采用 MIT 许可证。商用前请确认软件及模型权重的许可条款。

Stable Audio Tools 支持 Windows、Mac 和 Linux 吗?

Stable Audio Tools 支持:Linux, Windows。

运行 Stable Audio Tools 需要什么配置?

NVIDIA

如何安装 Stable Audio Tools?

1. Install the package. 2. Run the Gradio app from a clone of the repo with a pretrained model.

Stable Audio Tools 有哪些好的替代品?

常见替代品包括 AudioCraft (MusicGen), Bark, Whisper, whisper.cpp,详见本页对比表。