Advanced Text Comprehension
Consistent Voice Style Across Languages
Controllability and Security
Full Model Training Hours
100000
High-Fidelity Speech Synthesis
Mix-Language Comprehension
Multi-Speaker Voice Cloning
Open Source Model Training Hours
40000
Real-Time System Interaction
Sample Rate for Audio Output
24000
Studio-Level Online Editing
Ultra Low Latency
under 500 ms
Voice Customization Options
200