OCR in the Era of Large Language Models

A lightweight inference engine supporting speculative speculative decoding (SSD).

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

This is an open source project (formerly named Listen, Attend and Spell - PyTorch Implementation) for end-to-end ASR implemented with Pytorch, the well known deep learning toolkit.

Production-grade engineering skills for AI coding agents.

GLM-OCR: Accurate × Fast × Comprehensive

Fast and local neural text-to-speech engine

A fast, local neural text to speech system

A fast, local neural text to speech system

Contexts Optical Compression

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

VietASR - Vietnamese Automatic Speech Recognition

[NeurIPS'22] Squeezeformer: An Efficient Transformer for Automatic Speech Recognition

Open-Source Toolkit for End-to-End Speech Recognition leveraging PyTorch-Lightning and Hydra.

FSA/FST algorithms, differentiable, with PyTorch compatibility.

A Configurable template for a FastAPI application, with Authentication, User integration, Caching, Logging, Admin pages and a snappy CLI to control it all!

Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Andr…

Speech-to-text server framework with next-gen Kaldi

SoTA open-source TTS

SGLang is a high-performance serving framework for large language models and multimodal models.

Automatic Speech Recognition (ASR), Speaker Verification, Speech Synthesis, Text-to-Speech (TTS), Language Modelling, Singing Voice Synthesis (SVS), Voice Conversion (VC)

Port of OpenAI's Whisper model in C/C++

A mcp server to allow LLMS gain context about shadcn ui component structure,usage and installation,compaitable with react,svelte 5,vue & React Native

TTS Dia finetuning for Vietnamese

ViStreamASR - Real-Time Vietnamese Speech Recognition

The training program for libfacedetection for face detection and 5-landmark detection.

An open source library for face detection in images. The face detection speed can reach 1000FPS.