Just speak. Offline speech-to-text.

Added on February 1, 2026
Murmur

What is Murmur?

Turn your voice into text instantly with offline speech-to-text dictation for Windows — private, fast, and fully offline. Murmur is a powerful offline speech-to-text and voice dictation app for Windows, built on the OpenAI Whisper neural network. All voice processing happens entirely on your device. No internet connection, no cloud services, and no audio ever leaves your computer. Designed for productivity and privacy, Murmur lets you dictate text into any application using a single global hotkey. Just press the shortcut, speak naturally, and release — the transcribed text appears instantly at your cursor position so you can keep working without interruption. Murmur works everywhere: IDEs, terminals, browsers, chats, document editors, and enterprise applications. It is especially suited for corporate, offline, and NDA-restricted environments where cloud-based dictation tools are not allowed. Notebook is a PRO Feature Upgrade to PRO to unlock the Notebook feature and transcribe long recordings and files. * Transcribe audio and video files * Long recordings up to 3 hours * Export to TXT, PDF, DOCX, SRT * Unlimited saved transcriptions Key Features 🔒 100% Offline & Private All speech recognition runs locally on your PC. No uploads, no tracking, no telemetry, and no external servers. 🧠 Whisper AI Engine Accurate real-time speech recognition powered by OpenAI Whisper, supporting 99+ languages with automatic language detection. ⚡ High Performance GPU acceleration with NVIDIA CUDA and Vulkan for fast transcription. Automatically falls back to CPU when GPU is unavailable. ⌨️ Global Hotkey Dictation Start and stop dictation instantly from any application using a customizable system-wide hotkey. ✍️ Type Anywhere Transcribed text is automatically inserted into the active window or text field — no copy-paste, no app switching. 🗂 Local Transcription History All transcriptions are stored locally for browsing, editing, copying, and reuse. 🪶 Minimal Interface CleJu

Murmur's Core Features

100% Offline & Private

All audio processing happens locally on your PC — no cloud uploads, no telemetry, and no data leaves your device, making it suitable for NDA or GDPR-sensitive environments.

Global Hotkey Dictation

Start and stop dictation from any application using a customizable system-wide hotkey; transcribed text is inserted directly at the active cursor (no copy-paste or app switching).

Notebook (PRO) — File & Long-Form Transcription

Pro feature to transcribe long recordings and audio/video files (up to 3 hours), with exports to TXT, PDF, DOCX, and SRT and unlimited saved transcriptions.

Whisper AI Multilingual Engine

Powered by OpenAI Whisper with automatic language detection and support for 90+ languages for high accuracy across speakers and dialects.

High Performance with GPU Acceleration

Uses NVIDIA CUDA and Vulkan when available for fast, near-real-time transcription, and automatically falls back to CPU when needed.