VoxCPM2

VoxCPM2 is the latest release of OpenBMB's VoxCPM text-to-speech family. It is a 2B-parameter, tokenizer-free TTS model built on a MiniCPM-4 backbone and trained on more than 2 million hours of multilingual speech. It supports 30 languages, Voice Design (creating a new voice from a description), controllable voice cloning from reference audio, and 48 kHz studio-quality output. Weights and code are released under Apache-2.0 and are free for commercial use; models are on Hugging Face and ModelScope, with a live demo Space, documentation on Read the Docs and a technical report on arXiv.

A concise summary has not been added yet.

What is VoxCPM2?

VoxCPM2 is the latest release of OpenBMB's VoxCPM text-to-speech family. It is a 2B-parameter, tokenizer-free TTS model built on a MiniCPM-4 backbone and trained on more than 2 million hours of multilingual speech. It supports 30 languages, Voice Design (creating a new voice from a description), controllable voice cloning from reference audio, and 48 kHz studio-quality output. Weights and code are released under Apache-2.0 and are free for commercial use; models are on Hugging Face and ModelScope, with a live demo Space, documentation on Read the Docs and a technical report on arXiv.

How much does VoxCPM2 cost?

VoxCPM2 uses a Open source pricing model. Check the official website for current plans.

What is VoxCPM2 used for?

VoxCPM2 is listed on TolsAI under: AI Text-to-Speech, AI Voice Cloning, and Open Source AI Models.

Where can I use VoxCPM2?

VoxCPM2 is available on its official website, voxcpm.com.