OpenMOSS/MOSS-TTS
An open-source AI model family for high-fidelity, expressive speech and sound generation across diverse real-world applications.
Core Features
Detailed Introduction
MOSS-TTS Family is an advanced open-source suite of AI models developed by MOSI.AI and the OpenMOSS team, specializing in high-fidelity and highly expressive speech and sound generation. It is meticulously designed to address complex real-world scenarios, offering robust solutions for stable long-form speech, multi-speaker dialogue, intricate voice and character design, realistic environmental sound effects, and efficient real-time streaming Text-to-Speech (TTS). The project emphasizes versatile and efficient deployment, notably supporting PyTorch-free inference via llama.cpp and ONNX Runtime, making it a powerful and adaptable tool for diverse applications ranging from content creation and gaming to interactive systems and accessibility.