Data Collection and Analysis
Dental Appointment Booking
Emotion Tags
normalslowcryingsleepysighchuckle
Financial Product Guidance
Guided Emotion and Intonation
Hotel Reservation Assistance
Input Streaming for Lower Latency
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Multi-language Support
HindiTamilMultiple Indian Languages
Open Source Release Planned
Orpheus Speech Models
Medium (3B)Small (1B)Tiny (400M)Nano (150M)
Pretrained and Finetuned Models
Pretrained modelsFinetuned models
Sales Agent Functionality
Sample Finetuning Scripts
Sliding Window Detokenizer
Smart Conversation Management
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Suitable for Small Businesses
Supported Use Cases
RestaurantHotel BookingReal EstateLoan/Insurance/Credit CardRecruitmentISP SupportSalesDental BookingCustomer SupportAI Toys
Training Data Volume
100k+ hours of speech, billions of text tokens