- Home
- Text-to-speech tools
- Amphion
Amphion
What it is
Amphion is a free, MIT-licensed toolkit for generating audio, music, and speech, intended to support reproducible research and help junior researchers and engineers get started. Supported tasks include text-to-speech, singing voice synthesis, voice conversion, accent conversion, singing voice conversion, and text-to-audio; text-to-music is still in development. The toolkit includes diffusion, transformer, VAE, and flow-based architectures. Named text-to-speech models include FastSpeech2, VITS, VALL-E, NaturalSpeech2, MaskGCT, and Vevo-TTS; Vevo, FACodec, and Noro are listed for voice conversion. Vevo2 supports controllable speech and singing, including editing, style conversion, and melody control. Text-to-audio uses a latent diffusion model. Amphion also provides evaluation metrics for F0 and energy modeling, intelligibility, spectrogram distortion, and speaker similarity, plus interactive model visualizations. It unifies preprocessing for multiple open-source datasets and supports Emilia and Emilia-Pipe for in-the-wild speech data. Setup instructions use Python 3.9.15 with Conda or Docker; Docker requires NVIDIA drivers, NVIDIA Container Toolkit, and CUDA, and the dataset must be mounted into the container. WAV is listed as an export format.
For and against
For
- Free for research and commercial use under MIT.
- Supports several speech and singing generation tasks.
- Includes audio evaluation metrics and visualizations.
- Preprocessing supports Emilia and Emilia-Pipe.
Against
- Text-to-music is still in development.
- Docker setup requires NVIDIA drivers, toolkit, and CUDA.
- Dataset mounting is required for Docker setup.
The verdict
Amphion brings generation models, evaluation tools, and dataset preprocessing into a free toolkit. Text-to-music is not yet listed as supported, and Docker users need the specified NVIDIA components.
Who it suits
Amphion suits researchers and engineers working with speech, singing, or audio generation who want a toolkit that supports reproducible research. Its Linux and self-hosted platforms fit users prepared to follow its setup requirements.
Plans and prices
Amphion
MIT-licensed · free for research and commercial use
Read 5 Oct 2026 · github.com
On the record
- Audio evaluation
- Its evaluation metrics cover F0 and energy modeling, intelligibility, spectrogram distortion, and speaker similarity.github.com · read 5 Oct 2026
- Dataset preprocessing
- It unifies preprocessing for multiple open-source audio datasets and supports the Emilia dataset and Emilia-Pipe pipeline for in-the-wild speech data.github.com · read 4 Oct 2026
- Datasets
- Amphion unifies preprocessing for several open-source datasets and supports the Emilia dataset and Emilia-Pipe for in-the-wild speech data.github.com · read 5 Oct 2026
- Docker requirement
- The Docker instructions call for Docker, an NVIDIA Driver, NVIDIA Container Toolkit, and CUDA, and state that mounting the dataset with -v is necessary.github.com · read 5 Oct 2026
- Export formats
- WAVgithub.com · read 20 Sep 2026
- External model tools
- For content-based features, the README lists pretrained models including WeNet, Whisper, and ContentVec.github.com · read 5 Oct 2026
- Generation tasks
- Supported tasks include text-to-speech, singing voice synthesis, voice conversion, accent conversion, singing voice conversion, and text-to-audio; text-to-music is listed as in development.github.com · read 4 Oct 2026
- Installation
- The README documents installation with a setup installer using Python 3.9.15 and with a Docker image.github.com · read 5 Oct 2026
- Integrations and dependencies
- The project lists WeNet, Whisper, and ContentVec among pretrained models used for content-based features, and its Docker instructions require NVIDIA drivers, NVIDIA Container Toolkit, and CUDA.github.com · read 4 Oct 2026
- Intended users
- The project describes itself as supporting reproducible research and helping junior researchers and engineers enter audio, music, and speech generation research and development.github.com · read 5 Oct 2026
- License
- Amphion is released under the MIT License and is stated to be free for research and commercial use.github.com · read 5 Oct 2026
- License and cost
- Amphion is licensed under MIT and the README says it is free for research and commercial use.github.com · read 4 Oct 2026
- Models
- The toolkit implements diffusion, transformer, VAE, and flow-based model architectures.github.com · read 5 Oct 2026
- Purpose
- Amphion is an open-source toolkit for audio, music, and speech generation intended to support reproducible research and help junior researchers and engineers get started.github.com · read 5 Oct 2026
- Singing generation
- Vevo2 supports controllable speech and singing generation, including voice conversion, singing voice conversion, singing voice editing, singing style conversion, and melody control.github.com · read 4 Oct 2026
- Support and community
- The project invites contributions and links to a Discord channel for community engagement.github.com · read 4 Oct 2026
- Supported tasks
- The README marks text-to-speech, singing voice synthesis, voice conversion, accent conversion, singing voice conversion, and text-to-audio as supported, while text-to-music is in development.github.com · read 5 Oct 2026
- Technical limit
- Text-to-music is listed as in development rather than supported.github.com · read 5 Oct 2026
