Fully automatic censorship removal for language models

Code and data for the paper: IntentionQA: A Benchmark for Evaluating Purchase Intention Comprehension Abilities of Large Language Models in E-commerce (https://arxiv.org/pdf/2406.10173)

Real-Time VLAs via Future-state-aware Asynchronous Inference.

PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator (NeurIPS 2024)

Towards Human-Sounding Speech

Local UI to run and train LLMs and diffusion models, including Kimi K3, MiniMax-H3, Gemma 4, Qwen3.6, DeepSeek-V4, FLUX and more.

Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone

Command-line interface and Python library to transcribe pinyin to IPA. The tones are attached to the vowel of the syllable.

Target Speaker Extraction Toolkit

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

[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型

Genshin Datasets For SVC/SVS/TTS

video editing with vim/spreadsheet/sed/python. methodology inspired by BBC digital paper edit. "Excel-dit"

Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

A dataset of 222 digital musical scores aligned with 1068 performances (more than 92 hours) of Western classical piano music.

Foundational Models for State-of-the-Art Speech and Text Translation

Official implementation of "Separate Anything You Describe"

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

Book_7_《机器学习》 | 鸢尾花书：从加减乘除到机器学习；欢迎批评指正

🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.

A book about Text-to-Speech (TTS) in Chinese.

Towards hot directions in industrial end to end speech recognition

List of speech synthesis papers.

PyTorch Implementation of Google's Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data

an open-source implementation of sequence-to-sequence based speech processing engine

Crack LeetCode, not only how, but also why.

中文语音识别; Mandarin Automatic Speech Recognition;