VoxCPM
VoxCPM2 is an open-source text-to-speech system for generating multilingual speech, with voice cloning and voice design from a written description rather tha…
What VoxCPM is
VoxCPM2 is an open-source text-to-speech system for generating multilingual speech, with voice cloning and voice design from a written description rather than reference audio. It supports 30 languages and lists nine Chinese dialects. Output is 48 kHz audio, WAV is a listed export format, and reference audio is accepted at 16 kHz. The system provides a streaming generation API, and its repository reports a real-time factor as low as about 0.3 on an NVIDIA RTX 4090.
The project documents Python and command-line use, a local web demo, production serving, and on-device inference. Fine-tuning is available in full or through LoRA, with adaptation possible from 5–10 minutes of audio. The code and model weights use the Apache-2.0 license and are free for commercial use. The documented Python installation requires Python 3.10 or later but below 3.13, PyTorch 2.5.0 or later, and CUDA 12.0 or later. Voice design and controllable cloning results can vary between runs; unsupported languages may need testing or fine-tuning. The project prohibits impersonation, fraud, and disinformation, and recommends labeling generated content.
Good
- Supports 30 languages and nine listed Chinese dialects.
- Voice design works from a natural-language description.
- Full and LoRA fine-tuning are available.
- Apache-2.0 code and weights permit commercial use.
- Includes streaming generation and an OpenAI-compatible serving option.
Watch out
- Voice design and cloning results can vary between runs.
- Python installation requires specific Python, PyTorch, and CUDA versions.
- Unsupported languages may need testing or fine-tuning.
Who it suits
VoxCPM suits developers and teams seeking a free, self-hosted multilingual speech system with voice design or cloning. Its documented deployment options also address users who need API serving or on-device inference.
Verdict
VoxCPM2 offers multilingual speech generation, voice design, and multiple deployment paths under an Apache-2.0 license. Consider its installation requirements and the stated variability in voice results before choosing it.
Plans and prices
From VoxCPM's own pricing page, checked 8 Oct 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Free & Open Source | Free | No paid tiers or usage limits stated |
The facts
Each one read from VoxCPM's own pages, with the date it was checked.
What it does
- API access
- Yeschecked 20 Sep 2026
- API integration
- Its vLLM-Omni serving option exposes an OpenAI-compatible /v1/audio/speech endpoint.checked 8 Oct 2026
- Audio output
- The model outputs 48kHz audio and accepts 16kHz reference audio.checked 8 Oct 2026
- Commercial use
- The project states its code and weights use the Apache-2.0 license and are free for commercial use.checked 8 Oct 2026
- Deployment
- The repository documents Python and CLI use, a local web demo, production serving, and on-device inference.checked 8 Oct 2026
- Export formats
- WAVchecked 20 Sep 2026
- Fine-tuning
- It supports full and LoRA fine-tuning, with adaptation possible from 5–10 minutes of audio.checked 8 Oct 2026
- Integrations
- Documented serving options include vLLM-Omni with an OpenAI-compatible speech endpoint and Nano-vLLM for concurrent requests.checked 8 Oct 2026
- Languages
- VoxCPM2 supports 30 languages and lists nine Chinese dialects.checked 8 Oct 2026
- Limitations
- Voice Design and controllable cloning results can vary between runs, and the project says unsupported languages may require testing or fine-tuning.checked 8 Oct 2026
- Product
- VoxCPM2 is a tokenizer-free text-to-speech system that generates expressive multilingual speech.checked 8 Oct 2026
- Requirements
- The documented Python installation requires Python 3.10 or later but below 3.13, PyTorch 2.5.0 or later, and CUDA 12.0 or later.checked 8 Oct 2026
- Responsible use
- The project forbids using VoxCPM for impersonation, fraud, or disinformation and recommends labeling AI-generated content.checked 8 Oct 2026
- Streaming
- It provides a streaming generation API, and the repository reports real-time factor as low as about 0.3 on an NVIDIA RTX 4090.checked 8 Oct 2026
- Support
- The project links to Feishu, Discord, and its documentation wiki for community discussion and support.checked 8 Oct 2026
- Voice cloning
- It supports controllable voice cloning from a short reference clip and continuation cloning with reference audio and its transcript.checked 8 Oct 2026
- Voice design
- It can create a voice from a natural-language description without reference audio.checked 8 Oct 2026
The company
- Headquarters
- Beijing, Chinachecked 28 Sep 2026
- Maker
- The site identifies Modelbest Technology Co., Ltd. and offers custom solutions through [email protected].checked 8 Oct 2026
Money
- License and cost
- The model weights and code use the Apache-2.0 license, and the repository says they are free for commercial use.checked 8 Oct 2026
Where it runs
- Platforms
- web,macos,linux,api,self_hostedchecked 20 Sep 2026
- Platform support
- The README documents Apple Silicon MPS support and C++ inference on CPU, Metal, CUDA, or Vulkan.checked 8 Oct 2026
Alternatives to VoxCPM
- 1
Voicebox
Free planOpen sourceAndroidiOSWebAPI9.3From $12 a year - 2
Speechify
Free planAndroidiOSWebAPI9.0From $10 a month - 3
Resemble AI
Free planAndroidiOSWebAPI9.0From $280 a month - 4
iSpeech
Free planOpen sourceAndroidiOSWebAPI9.0Free plan - 5
Fliki
Free planAndroidiOSWebAPI8.8From $28 a month - 6
Voice.ai
Free planAndroidiOSWebAPI8.8From $5 a month