Escape the Keyboard: LocalWhisper Pro Helps You Actually "Touch Grass"
Описание и применение
LocalWhisper Pro позволяет диктовать текст на 150+ слов в минуту, автоматически транскрибировать, очищать и вводить его в активное окно без подключения к интернету, тем самым сокращая время за экраном и позволяя работать на открытом воздухе.
Задача
Users spend excessive time typing on screens, limiting outdoor activity and increasing screen time.
Результат
LocalWhisper Pro enables users to dictate at 150+ WPM, automatically transcribe, refine, and type text on‑device, cutting screen time and allowing outdoor work without internet or cloud dependencies.
Рабочий процесс
- 1. Capture audio locally with sounddevice.
- 2. Transcribe using faster‑whisper on CTranslate2 for up to 4x faster‑than‑real‑time output.
- 3. Refine transcript with local Ollama (llama3.1:8b) to remove fillers, summarize, or format code.
- 4. Inject refined text directly into the active window via pynput SendInput, bypassing the clipboard.
- 5. Continue dictation or exit to enjoy outdoor activities.
Доказательства
LocalWhisper Pro is an open-source, offline-first voice interface designed to minimize screen time.
Speech is up to 4x faster than typing. By handling filler-word scrubbing, formatting, and summarization automatically, users spend a fraction of the time staring at word processors and code editors.
Because both transcription and refinement run entirely on-device, you can take your laptop outside—onto the porch, into a garden, or down a hiking trail without cell signal.
It uses low-level hardware stroke injection (SendInput / pynput) rather than pasting through clipboard history, keeping workflows seamless and distraction-free.
Local speech recognition (faster-whisper / CTranslate2): Audio input is captured locally using sounddevice and processed through faster-whisper, a reimplementation of OpenAI's Whisper model running on the CTranslate2 inference engine.
- Дата записи: 2026-10-08T14:55:51.351Z