Emotion Tags
normalslowcryingsleepysighchuckle
Fast Transcription Speed
Most files within a few minutes
Guided Emotion and Intonation
High Accuracy Model
High (depends on audio quality)
Input Streaming for Lower Latency
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Multi-Language Transcription
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
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Supported Language List
EnglishSpanishFrenchGermanChineseand more
Training Data Volume
100k+ hours of speech, billions of text tokens