Conversational Speech Generation
Dataset Size
1 million hours
Human Interpretation Support
Model Sizes
Tiny: 1B backbone, 100M decoderSmall: 3B backbone, 250M decoderMedium: 8B backbone, 300M decoder
Multiple Sharing Methods
LinkInvite codeLocationQR-code
Multiple Speaker Handling
No Technical Knowledge Required
Objective Metrics
Word Error RateSpeaker SimilarityHomograph DisambiguationPronunciation Consistency
Partial Multilingual Support Planned
Planned for 20+ languages
Simultaneous Interpretation
Speech-to-Speech Translation
Subjective Metrics
Comparative Mean Opinion Score
Unlimited Channels
As many as needed
User Interface Language Options