Ant Group open-sources the all-modal large model Ming-Flash-Omni 2.0

Ant Group has open-sourced the all-modal Ming-Flash-Omni 2.0 large model, the industry’s first all-scenario audio unified generation model that can simultaneously generate speech, environmental sounds, and music within the same audio track. Users can give natural language commands to control timbre, speech rate, intonation, volume, emotion, and dialect. The model achieves an extremely low inference frame rate of 3.1Hz during inference, enabling real-time high-fidelity generation of minute-long long audio clips. (Science and Technology Innovation Board Daily)