Emotion Tags
normalslowcryingsleepysighchuckle
Guided Emotion and Intonation
Human Interpretation Support
Input Streaming for Lower Latency
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Multiple Sharing Methods
LinkInvite codeLocationQR-code
No Technical Knowledge Required
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
Simultaneous Interpretation
Sliding Window Detokenizer
Speech-to-Speech Translation
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Training Data Volume
100k+ hours of speech, billions of text tokens
Unlimited Channels
As many as needed
User Interface Language Options