DDN: A novel generative model with simple principles and unique properties. (ICLR 2025)

Speech Human Evaluation Estimation Toolkit (SHEET)

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.

Silero VAD: pre-trained enterprise-grade Voice Activity Detector

eSpeak NG is an open source speech synthesizer that supports more than hundred languages and accents.

✨ Split text by languages (e.g. 你喜欢看アニメ吗 -> 你喜欢看 | アニメ | 吗) for NLP tasks (e.g. parse, TTS). Powered by fasttext and budoux

Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

Ultimate Vocal Remover Inference CLI

[SIGGRAPH Asia 2024, Journal Track] ToonCrafter: Generative Cartoon Interpolation

FlashMLA: Efficient Multi-head Latent Attention Kernels

XiaoGua RTC Firmware is a firmware for ESP32 chip, which is designed by zideai.com.

Modern audio compression for the internet.

An MCP-based chatbot | 一个基于MCP的聊天机器人

A generative world for general-purpose robotics & embodied AI learning.

[ICCV 2025] SimVQ: Addressing Representation Collapse in Vector Quantized Models with One Linear Layer

A lightweight muji-moe chatbot created by Reecho.ai.

Acoustic Echo Canceller for Mobile Module Port From WebRTC

AEC3 Extracted From WebRTC

AI-based Audio Watermarking Tool

WebRTC Library for IoT/Embedded Device using C

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

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.

STM32 extension for working with STM32 and CubeMX in VSCode

Styled banners for your Readme made with html/css in SVG !!

AI powered speech denoising and enhancement

pkuseg多领域中文分词工具; The pkuseg toolkit for multi-domain Chinese word segmentation

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

Implementation for MatMul-free LM.