Open Source
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace.
Amical is an open-source, local-first AI dictation and note-taking app for macOS, Windows, and Android (with an iOS beta) that runs entirely on-device. It pairs OpenAI's Whisper for speech-to-text with local LLMs served through Ollama, turning spoken audio into clean, formatted text without sending anything to the cloud. The app detects which application is active and adapts its output accordingly, so dictating into an email, a Discord chat, an IDE, or a note reads naturally instead of like a raw transcript. Smart voice notes that produce summaries and structured tasks are still listed as in progress on the roadmap, alongside planned MCP integration and real-time meeting transcription, while hotkeys, voice macros, and custom workflows are already supported. It is built with Electron, TypeScript, Tailwind CSS, and Shadcn, and ships via a Homebrew cask, direct macOS/Windows downloads, and Google Play. The project markets itself as privacy-first: it works fully offline and lets users set up local models with a single click inside the app.
ggml-org
Pure C/C++ port of OpenAI Whisper for edge deployment
CJ Pais
A free, open-source, cross-platform speech-to-text app that transcribes your voice entirely offline — press a shortcut, speak, and have the text pasted into any app.
SYSTRAN
A CTranslate2-based reimplementation of OpenAI's Whisper that runs up to 4x faster at the same accuracy with lower memory, adding 8-bit quantization, batched inference, and word-level timestamps. MIT-licensed and FFmpeg-free.
m-bain
A fast open-source ASR pipeline on top of Whisper: word-level timestamps via forced alignment, 70x realtime batched inference, and speaker diarization (BSD-2).