Stars
Official repository of the paper "On-Policy Self-Distillation for Multi-Dialect ASR: Mastering Dialects, Retaining Mandarin".
DeepSeek Harness: Everything is a Plugin.
Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, P…
Open-Source Turn-Taking Detection Model and Dataset for Full-Duplex Spoken Dialogue Systems
The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.
An agentic skills framework & software development methodology that works.
We Speech Toolkit, LLM based Speech Toolkit for Speech Understanding, Generation, and Interaction
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
Fast and memory-efficient exact attention
Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.
Ongoing research training transformer models at scale
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Fun-ASR speech recognition models, with native Hugging Face Transformers support for Fun-ASR-Nano and separate FunASR, vLLM and llama.cpp deployment paths.
Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation
Voice Activity Detector (VAD) : low-latency, high-performance and lightweight
Open-source SenseVoiceSmall model for Mandarin, Cantonese, English, Japanese, and Korean ASR, language ID, emotion recognition, and audio event detection.
Official repository for the WenetSpeech-Chuan dataset.
A Large-scale Cantonese Speech Corpus with Multi-dimensional Annotation
Step-Audio 2 is an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation.
A native-PyTorch library for large scale M-LLM (text/audio) training with tp/cp/dp.
Open-source industrial-grade ASR models supporting Mandarin, Chinese dialects and English, achieving a new SOTA on public Mandarin ASR benchmarks, while also offering outstanding singing lyrics rec…
Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
easy-to-use implementation of the ISMIR 2013 Audio Degradation Toolbox
Label Studio is a multi-type data labeling and annotation tool with standardized output format