v0.1.136
- Fixed an issue where words spoken at a distance from the microphone were silently dropped from transcriptions.
- Fixed overlapping buttons in the recording interface when no microphone signal is detected.
Changelog
New updates and improvements to Koe.
v0.1.135
- Added a microphone level meter on the home screen with a target marker, so you can easily see if your mic is picking up enough sound.
- Added a warning when your microphone level is too low, prompting you to move closer to the mic for better results.
- Added a clipping warning when your microphone input is too loud.
- Improved the low mic level warning to appear in the overlay while dictating over other apps.
v0.1.134
- Improved the Voice Actions interface with a cleaner, more consistent design.
v0.1.133
- Added a sample console for testing and previewing Voice Actions.
v0.1.132
- Added Voice Actions, allowing you to set up custom commands that trigger automatically when specific words or phrases are detected.
v0.1.131
- Fixed an issue where the global shortcut to start dictation could fail to register correctly on certain keyboard layouts.
v0.1.130
- Improved recovery from unexpected crashes during active dictation, so your transcription is no longer lost.
- Fixed an issue where the app could get stuck in a bad state if a crash occurred while recording.
v0.1.129
- Improved reliability of long dictations so they are saved and recoverable if the app crashes or something goes wrong.
- Fixed an issue where valid long dictations were incorrectly rejected.
- Improved privacy protections during dictation sessions.
v0.1.128
- Improved the visual design of transcription and conversion screens.
v0.1.127
- Fixed a crash that could occur with audio devices when running under Rosetta.
- Fixed low-contrast text on the download prompt.
v0.1.126
- Bug fixes and improvements
v0.1.125
- Bug fixes and improvements
v0.1.124
- Bug fixes and improvements
v0.1.123
- Improved the desktop experience with a more polished and responsive interface.
- Added live validation to help catch errors as you type.
v0.1.122
- Improved app reliability by fixing issues that could leave recording in a broken state after errors.
- Reduced background CPU usage by checking for audio device changes less frequently.
- Fixed a memory issue where history could grow indefinitely over long sessions.
v0.1.121
- Bug fixes and improvements
v0.1.120
- Bug fixes and improvements
v0.1.119
- Bug fixes and improvements
v0.1.118
- Added Anthropic and Groq as bring-your-own-key (BYOK) AI providers for transforms.
- Fixed transforms using your own API key now returning clean text instead of raw data.
- Fixed the AI model picker dropdown being cut off in certain parts of the interface.
- Improved the API key setup area to clearly highlight when a key is missing and where it's needed.
v0.1.116
- Bug fixes and improvements.
v0.1.115
- Added Anthropic and Groq as bring-your-own-key provider options.
v0.1.114
- Bug fixes and improvements
v0.1.113
- Bug fixes and improvements
v0.1.112
- Bug fixes and improvements
v0.1.111
- Bug fixes and improvements
v0.1.110
- Bug fixes and improvements
v0.1.109
- Bug fixes and improvements
v0.1.108
- Bug fixes and improvements
v0.1.107
- Bug fixes and improvements
v0.1.106
- Bug fixes and improvements
v0.1.105
- Bug fixes and improvements
v0.1.104
- Bug fixes and improvements
v0.1.103
- Bug fixes and improvements
v0.1.102
- Fixed transcription quality degrading after repeated use by preventing CPU overcrowding between the transcription engine and other app components.
v0.1.101
- Improved transcription reliability by preventing slowdowns that could occur after repeated use.
v0.1.100
- Bug fixes and improvements
v0.1.99
- Improved voice detection accuracy to better distinguish speech from background noise.
- Fixed an issue where very long recordings could cause transcription to slow down or fail.
v0.1.97
- Added a Preferred Language setting to ensure transcriptions always come out in your chosen language, even when the app gets confused and outputs the wrong one.
- Fixed an issue where empty transcription results would not trigger the automatic restart, causing the app to get stuck after repeated failures.
v0.1.96
- Added a Preferred Language setting that ensures transcriptions are always output in your chosen language, fixing cases where speech in one language was incorrectly transcribed in another.
v0.1.95
- Fixed a delay at the start of your first recording after launching the app.
- Fixed an issue where audio could be lost during long recordings.
- Improved transcription reliability for long recordings (2 minutes and beyond).
v0.1.94
- Improved transcription reliability so no spoken audio is ever lost, even in long recordings.
- Fixed an issue where recordings over 90 seconds would have audio cut off and discarded.
- Improved accuracy for long dictation sessions by processing speech at natural pause boundaries.
v0.1.93
- Fixed the overlay status label to appear below the waveform and now shows 'Processing…' and 'Transforming…' during the appropriate stages.
- Fixed an issue where very long local recordings could cause slow or degraded transcription by capping audio at 90 seconds.
v0.1.92
- Improved the dictation overlay to show status labels below the waveform, with clearer "Processing…" and "Transforming…" indicators during each step.
v0.1.91
- Fixed an issue where transcription quality could silently degrade after extended use, causing the app to automatically refresh itself for a clean slate.
- Improved status indicator to only appear when there is a meaningful extra wait, reducing unnecessary visual noise.
v0.1.90
- Improved the recording status display to only show an explanatory label during the transcription processing step, reducing visual clutter.
v0.1.89
- Fixed a visual flash in the overlay that occurred between transcription and text transformation.
- Added a status label under the visualizer showing whether Koe is processing or transforming your speech.
- Improved transcription stability by ensuring the app fully restarts after updates, preventing quality degradation over time.
v0.1.88
- Fixed an issue where transcription quality could degrade after installing updates.
- Improved app restart behavior after updates to ensure a clean, fresh start every time.
v0.1.87
- Fixed an issue causing transcription quality to degrade after a few uses.
v0.1.86
- Improved local transcription accuracy by using smarter silence detection that handles whispers, varying volume, and background noise.
- Improved local transcription speed by reducing the amount of audio processed — silent portions are now stripped before transcription.
- Fixed an issue where repeated transcriptions would slow down and produce garbled text over time.
- Added a quality check that detects and discards garbled transcription output instead of pasting it, showing a clear error message instead.
v0.1.85
- Improved local transcription accuracy by better detecting and removing silence before processing speech.
- Fixed an issue where garbled or low-quality transcription results could be pasted into your app — these are now rejected with a clear error message instead.
v0.1.84
- Improved transcription accuracy for recordings with long pauses by automatically removing silence before processing.
- Fixed an issue where transcribing long audio clips (over 10 seconds) could return empty text.
- Fixed a crash that occurred when transcribing text containing accented characters (é, à, etc.).
v0.1.83
- Fixed a crash that occurred when transcribing text containing accented characters (such as French é, à, etc.), which also broke all subsequent transcriptions in the session.
- Fixed an issue where long audio recordings (over 10 seconds) returned empty transcriptions.
v0.1.82
- Bug fixes and improvements
v0.1.81
- Fixed transcription quality degrading after several recordings in a row.
v0.1.80
- Bug fixes and improvements
v0.1.79
- Fixed an issue where local transcription quality would degrade after several uses in a row.
v0.1.78
- Fixed transcription quality degrading over time during a session.
- Fixed opening files in your chosen code editor from the editor picker.
- Removed terminal apps from the editor picker, which would run files instead of opening them for editing.
v0.1.77
- Fixed an issue where opening files with an external editor could fail due to system permission restrictions.
- Removed terminal apps (such as Warp and Ghostty) from the editor picker, as they would try to run files instead of opening them for editing.
v0.1.76
- Fixed an issue where opening files with the editor picker could fail on some systems.
- Removed terminal apps from the editor picker, so only code editors and IDEs are shown.
v0.1.75
- Fixed the "Open hooks.json" button, which was previously blocked and did nothing.
- Improved the hooks settings layout for a cleaner, easier-to-use interface.
- Added automatic detection of installed code editors (Cursor, VS Code, Zed, Sublime Text, and more) with a dropdown to choose your preferred editor for opening hooks.json.
v0.1.74
- Fixed layout of the hooks settings panel, with dropdowns and inputs now properly arranged.
v0.1.73
- Added a Hooks management section in Settings to view, add, and remove hooks without editing files manually.
- Fixed a delay that caused voice-to-text to take several extra seconds to process.
- Fixed an error that appeared on every app launch related to API key setup.
- Improved error visibility in exported logs to make troubleshooting easier.
v0.1.72
- Fixed an error that occurred on every app launch related to API key migration.
- Fixed a delay of several seconds that affected text transformation after dictation.
- Improved error visibility in exported logs to help diagnose issues.
v0.1.70
- Fixed a 3-second delay that occurred before every transcription transform.
- Improved internal error logging to make issues easier to diagnose from exported logs.
v0.1.69
- Fixed an issue where text transformation could hang indefinitely after dictation.
- Fixed a bug where the app could silently fail on certain operations without any indication of what went wrong.
- Improved error visibility so that issues can be more reliably diagnosed from exported logs.
v0.1.68
- Fixed an issue where Koe could hang indefinitely when checking your account session, causing the app to become unresponsive.
v0.1.67
- Bug fixes and improvements
v0.1.66
- Fixed an issue where unplugging your selected microphone would cause an error — Koe now automatically falls back to your system default microphone instead.
v0.1.65
- Fixed an issue where AI text transformations would hang and never complete.
- Improved stability on macOS 26 Tahoe, reducing unexpected app crashes.
v0.1.64
- Improved stability on macOS 26 Tahoe, fixing crashes that could occur during normal use.
v0.1.63
- Added a custom dictionary feature that automatically corrects words and phrases in your transcriptions.
- Added transcript history, saving your past recordings so you can search and review them later.
- Improved transcription accuracy by using your dictionary words to guide the AI toward correct spellings and names.
v0.1.60
- Fixed an issue where Parakeet, Moonshine, and SenseVoice models would produce garbled or empty transcriptions after a few uses.
- Fixed the app freezing during model downloads.
v0.1.59
- Fixed an issue where empty transcriptions were being saved to history and auto-pasted.
- Fixed 'subscription required' errors that could appear after the app had been idle for a while.
v0.1.58
- Fixed an issue where the app could freeze on the loading screen.
v0.1.57
- Fixed an issue where the app would freeze on the loading screen.
v0.1.56
- Added a live voice transcription demo on the homepage — try it directly in your browser with no download required.
- Added copy-to-clipboard and Space bar hotkey for starting and stopping recordings in the demo.
- Fixed a bug where the model download progress bar would stay stuck at 0%.
- Added a compatibility check that warns you if your macOS version is too old to run Koe.
v0.1.55
- Added the ability to cancel an in-progress text transformation using the Escape key, the stop button on the overlay, or the Cancel button.
v0.1.54
- Added AI post-processing for transcriptions — clean up, format, summarize, translate, or apply a custom prompt to your dictated text after recording.
- Added post-processing controls to the system tray menu, showing active filters at a glance.
- Fixed several audio recording reliability issues, including lost words at the end of recordings, crashes when recording in quick succession, and quality degrading over time.
- Improved settings layout by moving API keys into a dedicated section and adding a link to the changelog.
v0.1.53
- Fixed audio recording reliability issues, including missing words at the end of recordings and crashes when recording multiple times in a row.
- Fixed an issue where a failed transcription engine could cause all subsequent transcriptions to fail until the app was restarted.
- Improved settings layout by moving API keys into a dedicated section, making the settings screen easier to navigate.
- Added a link to the changelog in settings.
v0.1.52
- Fixed an issue where cancelling a recording could fail or behave unexpectedly.
- Improved reliability of stopping a recording and getting your transcription quickly.
- Fixed the overlay window becoming unresponsive or stealing focus during recording.
- Improved handling of partial transcripts when a recording ends unexpectedly.
v0.1.51
- Added screenshot capture feature: assign a shortcut in Settings to capture your screen and process it through your custom hooks.
- Fixed several recording stability issues, including better handling of Bluetooth microphone disconnects, improved silence detection, and clearer error messages when something goes wrong.
- Fixed an issue where transcription could silently fail or produce incomplete text after long sessions or when the network lagged.
- Moved transcript history into the Settings panel, making it easier to find and optional to use.
v0.1.50
- Added a customizable hooks system — run shell commands or send web requests automatically at key points during recording and transcription.
- Added support for transcript middleware, allowing hooks to transform and rewrite transcribed text before it gets pasted.
v0.1.49
- Added optional noise suppression to reduce background noise during recordings, toggleable in Settings and the menu bar icon.
- Fixed an app freeze that could occur when starting or stopping recordings.
- Added the ability to stop or cancel an active recording from the menu bar icon and the main screen.
- Improved performance by keeping transcription models loaded in memory, reducing errors during heavy use.
v0.1.48
- Fixed an app freeze that could occur when starting or stopping recordings.
- Added noise suppression option in Settings and the menu bar to reduce background noise during dictation.
- Added the ability to stop or cancel a recording from the menu bar icon and the main screen.
- Fixed the overlay window blocking clicks in areas where it should be transparent, and improved its positioning when switching monitors.
- Added a warning during recording when no audio signal is detected, with a quick link to check microphone settings.
- Added a 3-second warm-up period for Bluetooth microphones before silence detection kicks in.
- Improved the update banner to show an 'Installing update...' status while a download is in progress.
v0.1.47
- Bug fixes and improvements
v0.1.46
- Added optional noise suppression to reduce background noise during recordings, toggleable in Settings and the menu bar.
- Fixed an app freeze that could occur when starting or stopping recordings.
- Added a short warm-up period before silence detection kicks in, improving compatibility with Bluetooth microphones.
- Fixed the overlay window blocking clicks on areas behind it when not in use.
v0.1.45
- Fixed an app freeze that could occur when starting or stopping a recording.
- Improved compatibility with Bluetooth microphones by adding a short warm-up period before silence detection activates.
- Updated the recording shortcut hint to correctly reflect whether hold or toggle mode is active.
v0.1.43
- Added automatic background update checks that notify you when a new version is available, with an in-app banner and updated tray menu.
- Fixed a bug where pressing Escape to cancel a recording could accidentally trigger actions in other focused apps.
- Fixed audio feedback sounds developing a delay over time during long sessions.
v0.1.42
- Added a system tray icon with interactive menus for switching models, microphones, recording mode, and settings — so you can control Koe without opening the main window.
- Added automatic background update checks with an in-app notification banner and tray menu item when a new version is available.
- Improved support for long recordings by compressing audio, allowing dictation sessions of up to ~25 minutes.
- Fixed an issue where warnings about missing API keys or subscription status would persist even after signing in or adding a key.
v0.1.41
- Added a system tray icon with interactive menus for switching models, microphones, recording mode, and settings — so you can control Koe without opening the main window.
- Improved support for longer recordings by compressing audio more efficiently, extending the maximum batch recording length to around 25 minutes.
- Fixed an issue where the 'no API key' warning would persist even after signing in or entering a valid key.
- Added a confirmation dialog when selecting a local model that hasn't been downloaded yet, showing the model size before you commit to downloading.
v0.1.40
- Improved support for longer recordings by compressing audio more efficiently, extending the maximum batch recording length to around 25 minutes.
- Fixed cloud model status not updating when switching between subscription and bring-your-own-key modes.
- Fixed streaming transcription authentication issues.
v0.1.39
- Added toggle recording mode — press once to start, press again to stop, as an alternative to hold-to-record.
- Added the ability to cancel any recording in progress by pressing Escape.
- Fixed the Globe/Fn key triggering the macOS emoji picker when used as a recording shortcut.
- Fixed text being lost during streaming transcription.
- Fixed a white flash on app startup.
v0.1.38
- Reorganized the app into a cleaner modular architecture for better performance and reliability.
- Fixed the overlay window appearing behind the loading screen on startup.
- Fixed layout and navigation issues across several screens.
v0.1.37
- Fixed verification code input so digits advance automatically, can be overwritten, and failed attempts let you cleanly re-enter a new code.
- Improved settings reliability so your preferences are no longer reset to defaults on startup.
- Fixed an issue where Pro subscription status could briefly appear incorrect after signing in or refreshing.
- Improved API key entry with a cleaner inline interface for bring-your-own-key providers.
v0.1.36
- Added auto-send option that automatically presses Enter after pasting your transcription.
- Added support for mouse button shortcuts, with a warning shown if third-party mouse software (e.g. Logi Options+) is intercepting the button.
- Fixed frozen waveforms and stuck recordings caused by quickly releasing the push-to-talk button.
- Fixed Enter key being interpreted as Cmd+Enter in Telegram and other apps after pasting, which prevented messages from sending.
- Fixed the macOS emoji picker appearing when using the Fn key as a shortcut.
- Fixed several login issues, including OTP verification hanging, Pro status not showing after login, and the verification code input not working reliably.
- Improved onboarding by combining Microphone and Accessibility permission steps into a single page.
- Reorganized Settings sections to put the most frequently used options first.
v0.1.35
- Added an auto-send option that automatically presses Enter after pasting your transcription.
- Added support for mouse button shortcuts, with a warning if third-party mouse software (like Logi Options+) is intercepting the button.
- Fixed auto-send not working in Telegram and other apps where Enter was being misinterpreted after pasting.
- Fixed frozen waveforms that could occur when releasing the push-to-talk button quickly.
- Fixed several bugs including OTP verification hanging, subscription status not updating after login, and the update checker requiring two clicks.
- Improved the permissions setup screen, combining microphone and accessibility into a single page.
- Improved the login code entry with individual digit boxes that auto-submit when complete.
- Suppressed the macOS emoji picker when using the Fn key as a shortcut.
v0.1.34
- Added an option to automatically press Enter after pasting your transcription, so dictated text can be submitted without any extra steps.
- Added a changelog page to keep you informed about what's new in each update.
v0.1.33
- Fixed an issue where signing out could fail if your session had expired.
v0.1.32
- Updated Privacy Policy and Terms of Service.
- Improved reliability by automatically retrying transcription if an error occurs, so your recording is not lost.
v0.1.31
- Fixed an issue where app updates could fail to download.
- Improved the overlay appearance and overall experience.
- Added the ability to delete individual entries from your dictation history.
- Added navigation support to toast notifications.
v0.1.30
- Improved the FAQ section with a smoother accordion interaction.
- Fixed an issue where app updates were not being detected correctly.
- Improved onboarding flow to automatically advance to the next step when the current one is completed.
- Updated privacy messaging for Pro mode.
v0.1.29
- Fixed an issue that could cause transcription results to appear delayed or out of order.
v0.1.28
- Added infrastructure for real-time streaming transcription.
v0.1.27
- Added an Upgrade to Pro option directly within the app.
- Fixed an issue where subscription status would not update correctly after completing a purchase.
- Improved the settings layout for cloud transcription models and API keys into a unified section.
v0.1.26
- Added an Upgrade to Pro option during onboarding.
- Fixed an issue where subscription status would not update correctly after purchasing.
v0.1.25
- Added a native macOS menu bar with quick access to About, Settings, and Check for Updates.
v0.1.24
- Improved dictation so it starts faster and no longer cuts off the end of phrases.
- Fixed an issue where closing the main window now hides it instead of quitting the app.
- Improved error messages to appear directly in the overlay.
v0.1.23
- Added live preview of transcribed text as you speak, shown directly in the overlay bubble.
- Improved permission setup flow, making it easier to grant the necessary access on first launch.
- Fixed auto-paste reliability so transcribed text is inserted more consistently.
- Improved overlay appearance with a translucent background and better error messaging.
v0.1.22
- Added real-time transcription that shows results as you speak.
- Fixed auto-paste not working reliably after dictation.
- Improved local transcription with support for multiple transcription engines.
v0.1.21
- Fixed an issue where macOS would prematurely ask for Input Monitoring permission before onboarding was complete.
- Improved the accessibility permission setup with a visual guide showing where to enable it in System Settings.
- Fixed the cloud model warning dialog appearing behind other UI elements.
v0.1.20
- Improved the experience for selecting subscription plans and transcription models.
v0.1.19
- Added sound effects for a more satisfying dictation experience.
- Improved transcription overlay with a cleaner toast-style display and the ability to cancel processing.
- Updated model selection to a convenient bottom drawer instead of a separate page.
- Redesigned history view with improved layout, clearer model names, and download progress indicators.
v0.1.15
- Improved app security protections.
v0.1.14
- Improved microphone permission handling with clearer feedback when access is needed.
v0.1.12
- Updated messaging and content around the Pro subscription.
v0.1.11
- Fixed a build configuration issue that could cause missing features in the previous release.
v0.1.10
- Added transcription history so you can review and copy your past dictations.
- Added support for multiple transcription providers, including Mistral, with a new model selection page and bring-your-own-API-key support.
- Improved onboarding experience with clearer permission steps, better animations, and a streamlined setup flow.
- Fixed auto-paste not working correctly when dictating into other apps.
v0.1.9
- Fixed a build packaging issue from the previous release.
v0.1.8
- Added Apple code signing and notarization, so macOS no longer warns about an unidentified developer.
v0.1.7
- Improved the macOS installer with a cleaner, more intuitive design.
v0.1.6
- Improved the visual appearance of the app header.
v0.1.5
- Fixed macOS installer packaging issues.
v0.1.4
- Initial release of Koe — voice-to-text dictation for macOS.