VoxCPM 2
Open-source multilingual TTS and voice cloning with controllable speech generation.
About VoxCPM 2
Overview
VoxCPM 2 is an open-source multilingual speech generation tool from OpenBMB and ModelBest. Its official demo highlights three workflows: Voice Design for creating a new voice from text instructions, Controllable Cloning for preserving timbre while steering style, and Ultimate Cloning for transcript-guided voice continuation. Official documentation also positions it as a developer-friendly stack with Python API, CLI, web demo, and deployment options.
Key Features
- Combines Voice Design, Controllable Cloning, and Ultimate Cloning in one stack.
- Supports 30 languages and documents dialect-oriented prompting for some Chinese varieties.
- Open-source and free to use, making it attractive for self-hosted or experimental workflows.
- Offers multiple access paths including Python API, CLI, web demo, and deployment guides.
AI no-code builder for apps, websites, and digital products
AI writing platform for marketing content
Cloud browser infrastructure for AI agents and web automation
AI video creation platform and Alibaba Cloud Model Studio video model family
AI voice generation platform for text-to-speech and voice cloning
Kanban app for orchestrating coding agents across isolated worktrees