Changelog

Browse and install transcription runtimes from the Runtime Manager in the activity bar. WhisperScript now supports MLX Audio on Apple Silicon Macs and ONNX Runtime on Windows, adding new local model options beyond whisper.cpp.

VibeVoice ASR is now available on Apple Silicon Macs through MLX Audio, with multilingual transcription and automatic language detection.

Parakeet TDT models are now available on both macOS and Windows. On macOS they run through MLX Audio; on Windows they run through ONNX Runtime.

When Zoom, Discord, Microsoft Teams, Slack, or FaceTime starts or ends, WhisperScript can show a notification with one-click actions to start or stop recording.

Record any app's audio alongside your microphone directly from the menubar window (Tray). Designed for quick, high-quality captures—great for interviews and meetings (Zoom, Discord, Chrome, etc.). What you can do: • Capture a specific active app (multiple selection supported) or your entire system output. • Record your microphone simultaneously. • Choose stereo or mono. • Save separate tracks (system + mic) and a mixed track for transcription. • Start/stop recordings instantly from the menubar. How to use: • Click the WhisperScript menubar icon to open the Audio Recorder. • Select your System Input (the app to capture) and Microphone. • Press Start Recording, then Stop to finish. Files are saved to your chosen location (defaults to Downloads). • Press Transcribe to import the finished recording into WhisperScript. Quality & reliability: • Drift- and delay-compensated dual-track recording keeps mic and system audio in sync—even during long sessions. • Supports 16-bit audio and sample rates from 8 kHz to 48 kHz. Permissions: • On first use, macOS will request Microphone and System Audio capture permissions. Please allow both. Availability: • macOS first; Windows support coming in a future release.

Added a settings button on hover for quick access to video options.

Added 'Close Video View' and 'Enter Picture-in-Picture' options to the video context menu.

Added a loading indicator during waveform generation.

Added 'Refresh Audio Waveform' action to the Transport Bar options.

Added 'Clear Waveform Cache' menu item to Help -> Debug.

Added support for CUDA toolkit v13 on Windows machines.

Added context menu actions to the search bar for direct removal of matching text or segment selection/deletion.

'Remove Text' command added to segment context menu.

Added support for MP4, MOV, and MKV files containing PCM audio streams.

Added buttons and shortcuts to jump to the previous/next search match.

Added support for 'Replace Current Match' in the search bar.

Added a 'Remove Line Breaks' command to the segment context menu to remove all line breaks in selected segments.

Added options to split regions, merge with previous/next regions, and delete regions.

Added helpful tooltips for advanced transcription parameters.

Added support for .webm video format.

Added new export formatting options that let you group transcript segments by Speaker or by Time Intervals (e.g. every 30 seconds), giving you more control over the structure and readability of exported transcripts. Also added an '+ Add All' button in the Export Manager to let you include all available files at once to your export list.

Added a Limit Characters per Line slider in the Transcription Configuration, which automatically inserts line breaks at the specified character limit.

Search now supports Regular Expressions and keeps a history of your recent searches, which you can browse using the up/down arrow keys. You can now use 'Replace All' to remove specific text by leaving the replacement field empty.

We've added a bunch of new shortcuts for subtitle editing and timeline navigation. You can now navigate between subtitles, trim or expand them at the playhead or by frames/milliseconds - all without touching the mouse! New: Split Region (Subtitle) at Playhead - just place the playhead and press `B` to open the Text Splitter popup.

You can now access a searchable list of keyboard shortcuts. Press `Shift + Ctrl + K` or go to `Help -> Open Keyboard Shortcuts Guide` in the menu.

Select segments and drag them out of the app while holding `Alt` to export them as media files. By default, adjacent segments are merged - configure via `File -> Export -> Drag Export Options -> Merge Adjacent Segments`.

- Global Keyboard Shortcuts Drawer - Split and trim regions flexibly: - Split at playhead or by text boundaries - Trim region start/end to playhead (`A`, `S`) - Expand to adjacent boundary (`Shift+A`, `Shift+S`) - Expand/trim by duration (`Alt/Meta + , . / m`) - Jump playhead to next/previous region with `Tab + Alt` - Customize nudge step in `Settings > Time Display`

View your transcript and video side-by-side, making it easier to follow along and edit in context.

Monitor subtitle quality in real-time (CPS, WPM, Word Count, Segment Duration) with customizable thresholds for ideal readability and pacing. Enable via: View -> Transcript -> Show Metrics. Configure thresholds: Transcript Toolbar -> Metrics Settings button.

Support for SMPTE timecode, ideal for film and broadcast workflows requiring frame-accurate time displays based on media frame rates. Click on the Time Display to open Time Display Settings.

Set a custom starting timecode for your project—perfect for syncing with video burn-in timecodes (e.g., if media starts at 10:00:00:00).

When using SMPTE mode, navigate by frames: • Move by 1 frame: Comma (,) and Period (.) • Move by 6 frames: M and Slash (/)

Enhanced subtitle editing capabilities: • Line breaks in subtitles are now fully supported. • Keyboard shortcuts to zoom waveform: Zoom-Out (R), Zoom-In (T).

Fine-tuned by Florian Zimmermeister, this model delivers up to 40% better accuracy in German transcription. A breakthrough for precision and speed.

Introducing Hebrew Whisper Large v3 and the memory-efficient Large v3 Turbo – state-of-the-art performance for Hebrew transcription with rapid results.

Toggle video Picture-in-Picture mode with the F11 key, allowing you to view videos outside the main app window.

Easily skip backward/forward by 10 seconds with dedicated keys and replay the current segment instantly using Shift+K.

Enjoy a few fun surprises hidden within the app – a little extra joy for our power users!

A refined option in Detect Speaker that trims non-speech at the segment edges for more accurate timing.

Stay informed with real-time update tracking directly within the download monitor.

Edit speaker tags on a per-segment basis to ensure precision in your transcript.

Right-click any segment to quickly assign speakers, delete, copy, merge, and more.

If you have a CUDA-enabled device, boost transcription speeds by enabling GPU acceleration in Advanced Options.

A guided setup to help new Windows users get started effortlessly.

Automatically separate speakers in transcripts with a simple click.

Easily track model downloads across the app interface.

Customize subtitle visibility, speaker lines, timecode placement, and video player size.

A guided setup designed to help new users dive right into WhisperScript with ease.

Customize the maximum number of speakers detected during transcription for tailored results.

Monitor model downloads from anywhere in the UI effortlessly.

Easily control subtitles, speaker lines, timecode position, and even resize the video player.

Run speaker detection after transcription to identify 'who spoke when' for up to 5 speakers. Rename speakers for consistent transcript tagging.

Get detailed explanations on why a file isn't supported – a smarter, more informative import process.

Experience complete feature parity with the Mac version – transcription, translation, search & replace, video support, and more.

Easily import media files on Windows with our intuitive drag & drop functionality.

A complete UI overhaul with a modern design, improved workflow, and a seamless multi-device experience.

Manage licenses and installations across devices with ease – now available for both Mac and Windows.

Transcribe videos (MOV, MP4, MKV) in-app with a redesigned toolbar for rapid access to editing actions.

Quickly edit text across your transcript with our intuitive search and replace functionality.

Export your transcripts to HTML, JSON, PDF, Word, and RTF with ease.

Access Finder, re-transcribe, or close files swiftly via the new filename dropdown.

Replaced the old integration with a new whisper.cpp backend for live transcription progress and detailed transcription time display.

Added an option in the transcription configuration to switch between CPU and GPU (Apple Silicon) processing.

For Pro users, introduced the 'small-distil-en' model with timestamps support and experimental improvements for medium and large distilled models.

WhisperScript now supports macOS Sonoma, along with 10%+ performance improvements on Apple Silicon devices.

Introducing three new Pro-only models to enhance transcription quality and speed.

Ensures that each model is fully downloaded and verified before use, preventing transcription errors from incomplete downloads.

Added a 'split on word' feature in the transcription window to ensure complete word rendering, especially in English.

A brand-new Transcription Settings Window with initial prompting, character limits per segment, and language translation to English.

New export options under the 'Transcript' tab, including current transcript, selection, and bookmarks.

New status bar, quick re-transcribe button, and updated auto-updater for a smoother public release experience.

Added support for AAC audio format and the ability to paste files from the clipboard directly into the app.

Added new buttons to select and bookmark all visible search results for a more efficient workflow.

Now you can subscribe to beta updates directly from Preferences, keeping you in the loop.

A new toggle to disable the re-transcribe button in the status bar when a tab is actively transcribing.