Search
On September 18, 2025, Memect AI, in collaboration with the Human-Machine Speech Interaction Lab at Tsinghua Shenzhen International Graduate School, proudly announced the launch of VoxCPM—a cutting-edge, 0.5-billion-parameter foundational voice generation model. This model has garnered acclaim for its exceptional performance across key metrics, including speech naturalness, timbre similarity, and prosodic expressiveness. Notably, it demonstrates unparalleled prowess in Zero-shot timbre cloning tasks, marking a significant leap forward in voice synthesis technology. To foster innovation and accessibility, VoxCPM is now available as an open-source resource on platforms like GitHub and Hugging Face, inviting developers and enthusiasts worldwide to explore and utilize its capabilities.
-
C114 Communication Network -
Communication Home
