Drive-Thru Voice Automation
Easy Scalable Installation
Emotion Tags
normalslowcryingsleepysighchuckle
Guided Emotion and Intonation
Input Streaming for Lower Latency
Integration Specialist Support
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Monthly Incremental Revenue Increase
6
Open Source Release Planned
Order Accuracy Improvement
Order Upselling Automation
Orpheus Speech Models
Medium (3B)Small (1B)Tiny (400M)Nano (150M)
Pretrained and Finetuned Models
Pretrained modelsFinetuned models
Sample Finetuning Scripts
Sliding Window Detokenizer
Staff Efficiency Optimization
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Superior Guest Experience
Training Data Volume
100k+ hours of speech, billions of text tokens