Atticus is a macOS voice assistant that uses hold-to-talk input and runs on top of Claude command-line mode. You press and hold a key to speak, and the transcribed request is passed to the assistant, which makes it a hands-free way to drive a coding or writing session. It is built for people who would rather talk than type when giving an agent instructions. The project is published as an open-source macOS application and relies on a local Claude setup instead of a separate hosted service.
Be the first to rate Atticus. Your feedback helps other users decide.
Speak instructions instead of typing them out
Drive a Claude coding session by talking
Turn spoken notes into text for an agent to act on
Similar tools in the same category — compare side-by-side
PDF to MP3 AI Connect turns documents into spoken audio using text-to-speech access that comes with ...
Nari Labs offers Qwen3-TTS and Qwen3-ASR as text-to-speech and speech-to-text APIs aimed at accuracy...
Sayso is dictation that runs on your own computer. You press a key, speak, and the text lands in the...
Guides and comparisons about audio tools
Eight voice, music, video and photo tools that keep media on your own machine: local dictation, an agent-driven DAW and actor-voice dubbing.
Eight AI tools launched in the last six weeks, from a Vulkan local model runner and SQLite agent memory to on-device budgeting and voice-driven coding.
AI audio tools for sleep stories in 2026: calming voiceovers, ambient soundscapes, and mixing tools that hold up across long tracks.
AI audio tools for audiobook production in 2026: natural narration, character voices, and mixing, with honest notes on consistency and licensing.
Free AI audio tools for podcasters in 2026: transcription, noise cleanup, remote recording, and AI narration that fit a zero-budget show.
