Acapela Group vs VoxCPM
Acapela Group
acapela-group.comVoxCPM
voxcpm.comSide by side
| Figure | Acapela Group | VoxCPM |
|---|---|---|
| Price from | €8.25/mo | Free |
| Free plan | — | Yes |
| Free trial | Yes | — |
| Open source | — | — |
| Web | Yes | Yes |
| Windows | Yes | — |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| iOS | Yes | — |
| Android | Yes | — |
| API | Yes | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 9 | 1 |
| Spec score · Text-to-Speech Software | 9.3 (#2) | 8.1 (#107) |
| Spec score · Text-to-speech tools | 9.2 (#9) | 8.8 (#24) |
| API access | Yes | Yes |
| Commercial use | Yes | The project states its code and weights use the Apache-2.0 license and are free for commercial use. |
| Deployment | Customer interaction products are available through cloud, sovereign cloud, on-premises, and embedded deployments. | The repository documents Python and CLI use, a local web demo, production serving, and on-device inference. |
| Headquarters | Mons, Belgium | Beijing, China |
| Languages | The company says its voices cover over 30 languages. | VoxCPM2 supports 30 languages and lists nine Chinese dialects. |
| Platforms | web,windows,macos,linux,ios,android,api | web,macos,linux,api,self_hosted |
| Support | Acapela says its team advises customers on voice selection and project requirements, and asks them to provide project, application, operating system, and language details. | The project links to Feishu, Discord, and its documentation wiki for community discussion and support. |
| Voice cloning | Yes | It supports controllable voice cloning from a short reference clip and continuation cloning with reference audio and its transcript. |
Plans
Acapela Group
- Yearly subscription€99 a year
- One-time license€999 a month
- Acapela Android voice€3.99, paid once
- Acapela TTS voices for Google Play€3.99, paid once
- My-Own-Voice yearly subscription€99 a year
- My-Own-Voice one-time licence€999, paid once
- Virtual Speaker€1,500, paid once
- Virtual Speaker first prepaid package€1,500, paid once
- Acapela TTS voices for NVDA€5,999, paid once
In their own words
Acapela Group
Acapela Group creates synthetic voices that turn written content into speech for services, apps, and devices. Its portfolio includes more than 250 voices in over 30 languages, with voice cloning, API access, and commercial use supported. Customer interaction products can be deployed through cloud, sovereign cloud, on-premises, or embedded setups. Acapela Cloud provides a web interface for creating and editing AI voice prompts and organizing files by project. Developers can use the Windows Server toolkit with proprietary APIs, SAPI5,.NET, C/C++, Java, MRCP v2, and W3C SSML. Rule-based pronunciation and lexicon tools allow specific terms to be tuned across languages. Acapela identifies voice agents and interactive voice response systems as use cases, and its NVDA voices serve blind and vision-impaired computer users. Platforms include Android, API, iOS, Linux, macOS, self-hosted, web, and Windows. The yearly subscription is €99 a year, billed excluding VAT; a three-month voice evaluation is included. Acapela is based in Mons, Belgium.
VoxCPM
VoxCPM is an open-source text-to-speech system whose VoxCPM2 model generates expressive multilingual speech. It supports 30 languages and lists nine Chinese dialects. Voice Design can create a voice from a natural-language description without reference audio, and voice cloning is supported. Output is 48kHz audio, reference audio is accepted at 16kHz, and WAV is listed as an export format. A streaming generation API is available; the repository reports a real-time factor as low as about 0.3 on an NVIDIA RTX 4090. Full and LoRA fine-tuning can adapt the model from 5–10 minutes of audio. Documented deployment includes Python and CLI use, a local web demo, production serving and on-device inference. vLLM-Omni exposes an OpenAI-compatible /v1/audio/speech endpoint, while Nano-vLLM supports concurrent requests. Apple Silicon MPS and C++ inference on CPU, Metal, CUDA or Vulkan are documented. Code and weights use Apache-2.0 and are free for commercial use. 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 cloning results can vary, and unsupported languages may require testing or fine-tuning. VoxCPM is headquartered in Beijing, China.