Advanced Emotion Control (Pro+)
Conversational Speech Generation
Cost Reduction
>95% cost reduction
Dataset Size
1 million hours
Demo Characters
Hades (Game NPC)Alex (Tutor)Dennis (Therapist)Ashley (AI Friend)Mark (Call Center)
Hosted, On-Premise, On-Device
HostedOn-PremiseOn-Device
Model Sizes
Tiny: 1B backbone, 100M decoderSmall: 3B backbone, 250M decoderMedium: 8B backbone, 300M decoder
Multiple Speaker Handling
Objective Metrics
Word Error RateSpeaker SimilarityHomograph DisambiguationPronunciation Consistency
Partial Multilingual Support Planned
Planned for 20+ languages
Realtime Streaming Output
Subjective Metrics
Comparative Mean Opinion Score
Templates for Realtime Pipelines