Dimmy.app

Dimmy.app is an AI voice dictation tool that turns speech into polished text across any app, built for knowledge workers who want privacy-first voice input on macOS and Windows. It processes audio locally by default using an offline Whisper Large v3 Turbo model (78 MB), with an advertised round-trip of about 0.4 seconds on M1 Macs, and only reaches the cloud if you opt into a provider like Groq, OpenAI, Deepgram, or Google Gemini. Beyond plain dictation, Dimmy has a Command Mode where you select text and give a spoken instruction such as make it more formal to edit it in place, and a Meeting Recaps feature that records calls, transcribes both sides, and produces structured summaries with decisions and action items. It supports 99 languages with auto-detection, ships 13 rewrite styles (Professional, Gen Z, Emoji, Custom, and others), works in 100+ apps including Slack, Notion, VS Code, Gmail, ChatGPT, and Claude, and lets you set per-app shortcuts and styles. For BestAIFor readers comparing dictation tools, Dimmy stands out for running transcription offline by default while still offering cloud models when you want them — a good fit for developers and writers who dictate across many apps and care about keeping audio on-device.

Dictating and rewriting text across desktop apps and recording meeting recaps with action items