Stars
SpaceXAI's coding agent harness and TUI. Fullscreen, mouse interactive, extensible.
AI-powered Claude skill for Spine 2D skeletal animation — auto-rig, animate, and preview characters
Official implementation based on MMOCR for paper "SegHist: A General Segmentation-based Framework for Chinese Historical Document Text Line Detection".
语音输入 + AI 润色，开源 Typeless 替代品。支持个人使用和团队/企业内部自部署。按下快捷键说话，文字自动输入到任何应用。
A local media library management tool powered by visual workflows
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively.
Official inference framework for 1-bit LLMs
MXNet implementation of RNN Transducer (Graves 2012): Sequence Transduction with Recurrent Neural Networks
eSpeak NG is an open source speech synthesizer that supports more than hundred languages and accents.
very fast speech-to-text, diarization, streaming (even in CPU) with NVIDIA Parakeet in Rust
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)
Finetune Nemo parakeet ASR model with new language (support 8 bit optimizer). Experimental birwkv-fastconformer TDT for long-form ASR(8.5 hours in single pass).
Interactive database view backed by CSV files, with rich column types, multiple views, sort & filter.
Qwen3-ASR is an open-source series of ASR models developed by the Qwen team at Alibaba Cloud, supporting stable multilingual speech/music/song recognition, language detection and timestamp prediction.
This repository primarily explores the use of Focal Codec to convert between speech signals and tokens, and employs an RNN model based on RWKV7 to achieve token prediction for speech generation.
Causal streaming adaptation of OpenAI Whisper for real-time transcription on small audio chunks.
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.
A calligraphy practice template primarily composed of grid-based "Tian Zi Ge" (Chinese character practice grids)
Open-source, self-hosted note-taking tool built for quick capture. Markdown-native, lightweight, and fully yours.
A fast, lightweight text-to-speech tool that runs entirely on your CPU. Give it text, pick a voice, and get a WAV file out.