Advanced Text Comprehension
Consistent Voice Style Across Languages
Enterprise Privacy Commitment
Expanded Model Support Planned
Human and Automated Safety Monitoring
Mix-Language Comprehension
Multi-Speaker Voice Cloning
No Training on Data Without Permission
Real-Time System Interaction
Reference Client Available
Streaming Audio Inputs/Outputs
Studio-Level Online Editing
Supports Text and Audio Inputs
Ultra Low Latency
under 500 ms
Voice Customization Options
200