AI voice

ASTRO — self-hosted, self-improving voice AI agent (PL): Raspberry Pi + Hailo NPU, wake word, STT/TTS, vision, LoRA self-training. Offline-first, open-core.

1M 以内大小的 macOS 语音输入工具

Local pipeline for narrated vertical business-story shorts: Z-Image stills, MiniMax H3 acting and a 2.5D paper camera compositor

Turn Claude Code / Codex into a multi-genre video studio: a video-as-code harness with 8 video types, 3 human review gates, a self-review loop, bilingual voice (Qwen3-TTS) and code-composed music & SFX.

Free, open-source dictation for macOS. Hold a key, speak, and your words appear at the cursor – transcribed locally with OpenAI Whisper.

A personal voice assistant that lives on your computer: Deep Agents + LangGraph, local speech, approvals, and skills she writes herself.

Phone Assistant for Codex: ask Codex to make a call, and an AI assistant handles it through your Mac's Phone app while you listen, join, or take over from the notch.

Skill do Claude Code: transcreve áudios da conversa com o Gemini 3.5 Transcribe (smart), com Whisper local de reserva

Companion code for Building Production Voice AI Agents (Book 5, Production AI Agent Engineering series)

Multilingual push-to-talk dictation for Omarchy and macOS (experimental). Choose your primary and secondary languages, or enable automatic language detection. Powered by ElevenLabs Scribe; transcribes and pastes into the focused app.

District AI for Linux: the native GTK4 + libadwaita client for District AI, written in Rust.

Hold Fn, speak, and your words appear wherever you type. A small macOS dictation app using Groq's Whisper.

Turns meeting voice notes into minutes, action items and Airtable records. Built with n8n, Groq, ElevenLabs Scribe and Telegram.

TypeFast 0.1.0 - Windows AI 语音输入法：按右Ctrl键说话，松开出字，需接入语音模型加润色模型

Voice-cast magic arena game. Speak or type incantations, powered by Jev, with local interpretation options.

🤖 Real-Time Voice Telemetry & Dynamic Audio Ducking Engine for AI Coding IDEs. Zero-silence agentic execution with JARVIS neural voice.

The open-source AI voice studio. Clone, dictate, create.

Local-first AI video dubbing with Whisper/WhisperX, Google translation, VoxCPM2 voice cloning, and optional LatentSync lip-sync.

Hand-drawn animated explainer videos made entirely with code — canvas JS visuals, synthesized music & SFX, AI voiceover, rendered with headless Chrome + ffmpeg.

Unofficial Stream Deck buttons and a hands-free 'Hey Claude' wake word for the Claude desktop app on Windows (AutoHotkey v2)