LocalScribe

English | Chinese (Simplified) | Japanese

Turn microphones, media files, and the sound playing on your Mac into editable text and subtitles—locally.

LocalScribe is a free, open-source native speech-to-text and audio/video transcription app for Apple silicon Macs, with no account required. It combines local recognition, floating live captions, subtitle editing and export, offline translation, long-task recovery, and on-device Gemma 4 transcript enhancement behind one SwiftUI interface. Audio, imported transcripts, and AI processing content are not uploaded by the app.

Current version: 1.7.0 (40) · Download DMG · First-launch installation guide · User documentation

[!TIP] New in 1.7.0 — original microphone audio: Optionally save the original recording for a new microphone transcription, play it back, and export an M4A alongside the transcript. Choose one of three quality profiles. Language changes now update menu titles, the AI prompt editor, and transcript export dialogs without restarting. Speech synthesis is disabled in this release.

See it in action

An eight-second loop of the home screen and editable transcript view, captured from the real 1.6.6 app with a non-private English sample. Images and GIFs include light and dark versions that follow your viewing theme.

[!IMPORTANT] Apple SpeechAnalyzer recognition and floating live captions require macOS 26. The app itself supports macOS 15.5 or later. On macOS 15.5–25, manually select Whisper, SenseVoice, or Parakeet on the home screen. SenseVoice and Parakeet currently support file transcription only.

Download and install

1. Download the latest DMG.
2. Open it and drag LocalScribe to Applications.
3. Try opening the installed app once. If macOS blocks it, close the alert, then open System Settings → Privacy & Security → Security. Find this app’s warning and choose Open / Open Anyway, then authenticate or confirm as requested.

Installing for the first time on macOS 26 or 27? Follow the step-by-step first-launch guide, including what to do if Open Anyway is missing.

For illustrated steps, troubleshooting, and SHA-256 verification, see the Download and Installation Guide.

[!WARNING] This build has an ad-hoc integrity signature only. It is neither Developer ID signed nor notarized by Apple. Override the warning only if you trust this repository and its Release. SHA-256 files are included with every Release.

On-device Gemma 4 transcript enhancement (new in 1.6.4)

Transcription is only the starting point. After transcription or transcript import, open AI Enhancement in the right inspector to turn the result into a cleaner, deliverable document without copying it to another AI service or uploading it.

Proofread and refine with a live diff

Gemma removes filler, repetition, and obvious transcription errors while preserving the intended meaning. Original and suggested text update in a live diff alongside batch progress, model status, and the on-device privacy indicator. Segments that fail validation keep their original text.

Distill long transcripts into focused summaries

Summaries replace the text preview while keeping an undo snapshot of the original. Add one-time instructions for names, terminology, writing style, summary length, priorities, or format, and optionally save reusable instructions in Settings. AI output should still be reviewed.

The default Gemma 4 E2B IT Q4 model is about 2.8 GB; an optional E4B model of about 4.6 GB can be enabled in Settings. Models are downloaded on demand, verified, and run on the Mac through llama.cpp and Metal. On Macs with 8 GB of physical memory or less, AI Enhancement is off by default and can be enabled manually in Settings after confirming the memory warning. Gemma starts loading only after you select Proofread or Summarize. Completing a transcription or opening its result never preloads Gemma. Before loading Gemma, LocalScribe releases its active recognition and NLLB translation runtimes; the Gemma helper exits when the task finishes.

Why LocalScribe?

• Local by default. Recognition and optional translation run on the Mac; input audio and imported transcripts are not uploaded by the app.
• More than dictation. Use microphone input, media files, Mac audio, floating live captions, transcript editing, recovery, and subtitle export in one workflow.
• Multiple offline engines. Choose Apple Speech, whisper.cpp with Metal, SenseVoice, or NVIDIA Parakeet instead of being locked to one model family.
• Long-task workflow. Progressive results, append-only recovery journals, bounded transcript rendering, pause/resume, and resumable sessions are designed for long recordings.
• Editable deliverables. Import, search, replace, trim, translate, and export TXT, Markdown, JSON, PDF, SRT, or WebVTT.
• On-device AI transcript enhancement. Use Gemma 4 to proofread, refine, or summarize a transcript, with custom instructions for terminology, style, focus, and format.
• System-aware interface. English, Simplified Chinese, and Japanese switch automatically with the Mac language preference, using an extensible localization structure.

Transcript editing and translation

The screenshots are from version 1.6.6 (37) of the real macOS app and use non-private English sample text. Both appearances are captured from the app, without the Computer Use pointer. LocalScribe includes complete English, Simplified Chinese, and Japanese interfaces.

Save original microphone audio

For a new microphone transcription, enable Save Original Audio in the settings sidebar before starting. This option is off for each new task and becomes locked once recording starts. Storage Priority, Quality Priority, and Highest Quality produce M4A files using AAC or ALAC. The app adapts to the microphone format automatically.

The recording keeps silence and other captured sounds independently of recognition. Pause gaps are omitted. Once saving finishes, use play/pause, seeking, and 0.5×, 1×, or 2× playback. Export the original recording alongside TXT, Markdown, JSON, PDF, SRT, or WebVTT with the same base filename. Editing or translating the transcript does not change the original audio.

Original-audio saving is unavailable for media-file transcription, Mac system audio, floating captions, and microphone transcription appended to an imported transcript. If audio saving fails, recognition can continue; any recoverable partial audio is labeled. The latest recoverable task retains its audio locally. Speech synthesis and its settings are disabled in this release.

Languages

LocalScribe follows the preferred language order in macOS by default. You can also switch the app immediately among English, Simplified Chinese, and Japanese in Settings without changing the system language; the menu bar and live-caption source controls update at the same time. Recognition-language menus place English and the Mac's language in a deduplicated Recommended Languages section. Settings also provide System, Light, and Dark appearance choices, plus current microphone, speech-recognition, and system-audio permission status.

Recognition and translation engines

Prefer Apple's built-in engines whenever possible. If your macOS version and languages are supported, use Apple Speech for transcription and Apple Translation for translation. Their native optimization and output quality are noticeably better than those of the third-party transcription and translation models available in LocalScribe.

Whisper file transcription uses the model's internal sliding windows rather than fixed, non-overlapping application chunks. For longer media, the bundled Silero VAD can skip silence while retaining speech padding and overlap. Output filtering considers silence, confidence, repetition, and known hallucination patterns.

Advanced transcription controls

For supported third-party engines, the right inspector includes an Advanced section that is collapsed when a task opens. Whisper exposes a model prompt, temperature and fallback controls, beam or greedy search settings, context length, silence and confidence filters, VAD, and a guarded automatic thread policy. SenseVoice and Parakeet expose the options their local runtimes actually support. File-only and microphone-only settings appear only for the matching source.

Numeric options support both direct keyboard entry and sliders or steppers, with range validation and short hover explanations. The app remembers the last engine, model, and advanced configuration for the next task; Restore Model Defaults resets only the selected engine.

Highlights

• Apple Speech live captions for microphone, Mac audio, or mixed input on macOS 26.
• whisper.cpp Metal inference with bundled Silero VAD and automatic CPU fallback.
• Downloadable Whisper, SenseVoice, and Parakeet model choices.
• Adaptive progressive text display for file and microphone transcription.
• Editable transcripts with search, replacement, selection deletion, and range trimming.
• SRT, WebVTT, TXT, Markdown, and LocalScribe JSON import.
• TXT, Markdown, JSON, PDF, SRT, WebVTT, and clipboard export.
• Apple Translation by default, with optional local NLLB INT8 translation.
• Append-only session journals and recovery snapshots for long-running work.
• A command-line interface for models, transcription, export, and translation.

Requirements and current limitations

• macOS 15.5 or later.
• Apple silicon (arm64); Intel Macs are not supported.
• Apple SpeechAnalyzer recognition and live captions require macOS 26.
• On macOS 15.5–25, select Whisper, SenseVoice, or Parakeet manually. SenseVoice and Parakeet currently support file transcription only.
• Live-caption translation is currently disabled; post-transcription translation remains available.
• Gemma 4 AI Enhancement is off by default on Macs with 8 GB of physical memory or less; Settings allows manual activation after confirmation. Models are downloaded on demand and loaded only after you select an AI feature.
• Local NLLB translation is unavailable below 4 GB of physical memory. The check allows a 128 MiB tolerance around the threshold for system-reported memory. Apple Translation remains available.
• Public Developer ID signing and Apple notarization are still pending.

Microphone input requires microphone permission. Capturing Mac audio requires Screen & System Audio Recording permission. Apple Speech and Apple Translation may download language assets managed by macOS.

Build from source

Install Xcode and its Command Line Tools. The repository includes the native runtimes required by the app; large recognition, NLLB, and Gemma 4 models are downloaded only when selected.

ruby generate_project.rb

xcodebuild \
  -project LocalScribe.xcodeproj \
  -scheme LocalScribe \
  -destination 'platform=macOS,arch=arm64' \
  build

Run the test suite:

xcodebuild \
  -project LocalScribe.xcodeproj \
  -scheme LocalScribe \
  -destination 'platform=macOS,arch=arm64' \
  test

CI runs the tests and Release static analysis on GitHub's macOS 26 Apple silicon runner.

The local packaging script creates ZIP and DMG artifacts and validates nested signatures, architectures, deployment targets, embedded Mach-O files, archive extraction, DMG mounting, and helper startup:

./tools/package_local_release.sh

By default it uses the configured local certificate. Set CODESIGN_IDENTITY=- to make the same ad-hoc package produced by GitHub Actions. Pushing a tag that matches the version in Info.plist (for example, v1.7.0) or includes the build number (for example, v1.7.0-build40) runs release-unsigned.yml, verifies the package, and creates the GitHub Release without storing a certificate or password in GitHub Secrets. Developer ID signing, timestamping, notarization, and stapling remain the preferred public distribution path.

CLI

LocalScribe.app/Contents/MacOS/LocalScribe --cli help
LocalScribe.app/Contents/MacOS/LocalScribe --cli models --json
LocalScribe.app/Contents/MacOS/LocalScribe --cli transcribe input.mp4 \
  --engine whisper --language en_US --format srt --output output.srt

Privacy and updates

LocalScribe does not upload recognition audio, imported transcripts, or content processed by Gemma. Recognition, translation, and AI transcript enhancement run on the Mac. Network access is used for on-demand recognition, translation, and Gemma model downloads, automatic or manual update checks and downloads, and opening external documentation. Automatic updates can be disabled in Settings.

The built-in updater checks at launch and every six hours by default, reads update.json from the latest GitHub Release, downloads its ZIP, verifies SHA-256, checks the bundle identifier and version, and asks before replacing and relaunching the app. This update path is not a substitute for a notarized public release.

Documentation

• User documentation
• Download and Installation Guide
• Acceptance and regression notes
• SherpaOnnx build notes
• Roadmap
• Contributing
• Report an issue
• Media and recommendation kit

License

LocalScribe source code is available under the MIT License. Third-party components and models retain their original licenses; see THIRD_PARTY_NOTICES.md and the license files under Vendor.

The optional NLLB model is distributed upstream under CC-BY-NC-4.0.