FlashLabs-AI-Corp/FlashLabs-Chroma

Worlds first open-source real-time end-to-end spoken dialogue model with personalized voice cloning.

50
/ 100
Established

Built on a hybrid architecture combining Qwen2.5-Omni for reasoning with Llama3-based backbone and decoder layers, Chroma processes raw audio directly and generates synchronized text-speech outputs using Mimi codec at 24kHz sampling. The model supports zero-shot voice cloning by conditioning generation on reference audio prompts, enabling style transfer without task-specific fine-tuning, and integrates with Hugging Face transformers for seamless PyTorch inference with bfloat16 quantization support.

545 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 13 / 25
Community 17 / 25

How are scores calculated?

Stars

545

Forks

59

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jan 28, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/FlashLabs-AI-Corp/FlashLabs-Chroma"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.