Havuz

CueFlow (随读) — a notch-adjacent macOS teleprompter that listens as you read and auto-scrolls to follow you. 会听你朗读、自动跟随滚动的 Mac 刘海提词器。

A blazing-fast, offline push-to-talk voice assistant. Features real-time Whisper.net transcription, auto-typing under caret, smart VAD, and

agent-oriented wer open metrics

Hackable voice AI assistant for Windows: wake it with your voice, bring your own LLM (Claude/Groq/Ollama/OpenAI/Gemini), and it actually doe

Windows 线上会议面试助手：多会议平台路由、本地流式语音识别、可选领域模块，并针对金融与会计场景做专项优化。

本地优先的直播精彩切片工作流：Qwen3-ASR 全时间轴转录，Agent 筛选与人工确认，Remotion 9:16 渲染。

Chrome extension that turns video subtitles and tab audio into real-time 3D Chinese Sign Language with local Whisper.

Speech and language toolkit for Obsidian

ThorOS - a voice-controlled, AI-first Linux distribution built on Debian 13. Say what you want; on-device AI plans it and permissioned agent

Streaming on-device speech recognition for Android — NEON-accelerated FastConformer (32M params), cache-aware streaming with 80 ms look-ahea

Accepted at ISCSLP 2026 · Target-level benchmark for raw-input Chinese news TTS pronunciation

Run NVIDIA's tri-modal Nemotron Omni (text+vision+audio) entirely on Apple Silicon — pure MLX runtime, parity-verified against NVIDIA's refe

Easy access to speech data across 142 African languages for training TTS and ASR models.

Fcitx5 voice input addon — PulseAudio/PipeWire capture, Silero VAD, OpenAI-compatible & Volcengine Doubao streaming ASR, AsrSession architec

Speech-first inference runtime in Rust — TTS / ASR / speech-to-speech / VC / speaker ID / VAD. An ONNX Runtime alternative that loads GGUF &

Privacy-first dual-channel meeting transcription & AI summary for macOS — 100% local, powered by whisper.cpp and Ollama

中英混著講，落下的是乾淨的繁體中文。macOS 選單列語音輸入，whisper.cpp 全程在本機跑，kubectl、Terraform 這類術語內建修正。

Streaming on-device speech recognition for iOS — NEON-accelerated, encrypted FastConformer (32M params), RTF 0.08–0.10 on iPhone 13 Pro Max.

Real-time speech translation with transparent on-screen subtitles. 100% local, no cloud

面向杭电「上课啦」的 Windows 课堂辅助工具，支持本地语音识别、关键词提醒与可选签到辅助。

Голосовой ИИ-ассистент для Linux/macOS: wake word, LLM + bash-агент, Telegram-бот, GUI. Бесплатно и локально.

Native macOS push-to-talk dictation app — hold key, speak, release to transcribe and paste text via ChatGPT Web or Groq API; no API key requ

Whisper-large-v3 fine-tune for Indonesian + Javanese + Sundanese ASR. Lhotse + k2 lattice rescoring pipeline.