A lightweight, powerful framework for multi-agent workflows

An neural full-band audio codec for general audio sampled at 48 kHz with 7.5 kps or 4.5 kbps.

Source & evaluation code for ICAMCS 2024 paper "Emotional Vietnamese Speech-Based Depression Diagnosis Using Dynamic Attention Mechanism"

🐫 CAMEL: Finding the Scaling Law of Agents. The first and the best multi-agent framework. https://www.camel-ai.org

AudioBench: A Universal Benchmark for Audio Large Language Models

A Conversational Speech Generation Model

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Get up and running with Llama 3.3, DeepSeek-R1, Phi-4, Gemma 3, and other large language models.

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.

A high-throughput and memory-efficient inference and serving engine for LLMs

End-to-end stack for WebRTC. SFU media server and SDKs.

Examples and guides for using the OpenAI API

A repo containing download guidance and corresponding scripts of the VoxBlink dataset.

LLaMA-Omni is a low-latency and high-quality end-to-end speech interaction model built upon Llama-3.1-8B-Instruct, aiming to achieve speech capabilities at the GPT-4o level.

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

A fast multimodal LLM for real-time voice

Build datasets using natural language

Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key

KUIELAB-MDX-Net got the 2nd place on the Leaderboard A and the 3rd place on the Leaderboard B in the MDX-Challenge ISMIR 2021

zero-shot voice conversion & singing voice conversion, with real-time support

SSL Layerwise analysis for speech deepfake detection

https://hf.co/hexgrad/Kokoro-82M

LLaSA: Scaling Train-time and Inference-time Compute for LLaMA-based Speech Synthesis

The official Pytorch implementation of "Frame-wise streaming end-to-end speaker diarization with non-autoregressive self-attention-based attractors". [ICASSP 2024] and "LS-EEND: long-form streaming…

Unified automatic quality assessment for speech, music, and sound.

Score-based Generative Models (Diffusion Models) for Speech Enhancement and Dereverberation

Collection of papers on state-space models

Gradio WebUI for creators and developers, featuring key TTS (Edge-TTS, kokoro) and zero-shot Voice Cloning (E2E, F5-TTS, CosyVoice), with Whisper audio processing, RVC voice changer, YouTube downlo…