VoxCPM
Fast and high-quality speech cloning model
VoxCPM: Tokenizer-Free TTS for Context-Aware Speech Generation and True-to-Life Voice Cloning The model achieves an excellent balance between speed and realism, with short sentence generation speed reaching real-time, and supports emotion control (generated speech is similar to the emotion of the reference audio), polyphonic characters, and other controls, as well as fine-tuning. The AI Pod provides an online web-ui generation interface and API for VoxCPM, which is recommended for use in scenarios requiring realistic voice cloning. For faster long-sentence speech generation, please refer to the AI Pod documentation to use the built-in speech generation model. In short: This is the strongest synthesized speech and cloned voice model of 2025, and it's enough.
