Background Noise Handling
Emotion Tags
normalslowcryingsleepysighchuckle
Guided Emotion and Intonation
Industry Solutions
HealthcareLegal & Law EnforcementFinancial ServicesHospitality & TourismTechnology & EngineeringReal Estate
Input Streaming for Lower Latency
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Open Source Release Planned
Orpheus Speech Models
Medium (3B)Small (1B)Tiny (400M)Nano (150M)
Pretrained and Finetuned Models
Pretrained modelsFinetuned models
Sample Finetuning Scripts
Sliding Window Detokenizer
Smart Transcription Accuracy
99.9%
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Supported Export Formats
Multiple formats
Supported Language List
98
Total Transcription Hours
125000
Training Data Volume
100k+ hours of speech, billions of text tokens