Changelog

New features

• Meeting recording with on-device speaker diarization — capture mic and system audio together, and every segment is automatically labeled with who said it. Diarization runs 100% locally: voices never leave your device. System audio capture now works on Windows and Linux (Chromium loopback and the native audio portal), not just macOS.

• Local live transcription preview — see your words appear in real time as you speak, streamed from a local Whisper or Parakeet model before you release the hotkey. Fully on-device, no cloud round trip.

• Self-hosted support — first-class option to point OpenWhispr at your own LAN server for both transcription and agent inference. One click to swap between OpenAI, Anthropic, Gemini, a local GGUF, or your self-hosted endpoint.

• Speaker to contact linking — voice profiles bind to calendar attendees, so recurring participants get named automatically after the first tag.

• Improved echo cancellation — WebRTC AEC now runs as a native sidecar, paired with a custom cross-correlation detector and text-level dedupe to cleanly handle external speakers, system-audio capture, and low-echo-quality mics.

• Gemma 4 models — Gemma 4 31B and Gemma 4 26B MoE added to the local model registry.

• Smart transcription error handling — when a transcription fails, inline CTAs let you retry, switch provider, or open logs.

Linux

• Hyprland global hotkeys via hyprctl, plus correct active-window detection for terminal paste.

• GNOME shortcut conflict detection, active-key display, and proper fallback when the requested shortcut is taken.

Fixes

• Windows: installer DPI scaling corrected, GPU compositing re-enabled.

• Local Whisper: stricter speech gate cuts hallucinated output on silent audio.

• Diarization: session-scoped results, disk spooling for long meetings, reliable spinner, correct speaker label propagation.

• Notes: transcripts scope to their originating note, view auto-switches when a meeting recording starts.

• UI: fixed 420px window width that grows only with text, consistent across phases; preview overlay redesigned.

• Settings: inference mode auto-switches when you select a new provider.

• i18n: speaker labels, cloud settings, integrations, and calendar strings tightened across locales.

• Stability: 22 React Compiler lint issues resolved, tighter effect dependencies, cleaner hook usage across 15 files.

Rich Text Notes

• Notes now use a rich text editor with Obsidian-style live preview — Markdown syntax hides as you type, giving you a clean writing experience

Meeting Transcription Upgrades

• Dual-channel transcription — mic and system audio are captured separately with speaker-labeled chat bubbles so you can see who said what

• Timestamped segments — meeting transcripts now include timestamps in chronological order

• Smarter meeting notes — AI-generated summaries are now speaker-aware for more accurate meeting recaps

New Local Models

• Mistral Nemo 12B and Gemma 3 12B are now available for on-device AI processing

macOS Audio Improvements

• Native system audio capture via CoreAudio Tap — no more "screen recording" permission prompt on macOS 14.2+

• macOS 15+ now shows the correct system audio consent dialog instead of the legacy screen recording one

Linux

• KDE Wayland support — native global shortcuts now work on KDE Plasma via D-Bus, joining GNOME and Hyprland support

• Fixed KDE Plasma overlay window and hotkey behavior

• Fixed clipboard paste reliability on KDE Wayland

Simplified Permissions

• Permission prompts consolidated to a single "Grant Access" button

• Permissions are now re-checked against the OS each time you open settings, so the UI always reflects your actual state

Bug Fixes

• Fixed Gemini agent streaming routing to the wrong endpoint

• Fixed Windows mic volume being permanently altered during dictation

• Fixed mono transcription failure on Linux

• Fixed Bluetooth audio issues during meetings

• Fixed meeting detection notifications firing when you're already in a meeting

• Fixed held modifier keys not releasing before paste on Windows

• Fixed paused media being unpaused during dictation

• Fixed Google OAuth users skipping onboarding

• Fixed AI cleanup prompt refusing to transcribe command-like speech

• Fixed agent hotkey conflicts not showing a warning

• Meeting detection improvements

Meeting Mode

• Meeting mode hotkey — Assign a dedicated hotkey to instantly snap the panel into meeting mode and start a new meeting note

• Smarter meeting detection — Reduced false positives; background apps like FaceTime no longer trigger notifications unless mic activity is detected

• Meeting detection toggle — Disable meeting detection entirely from Settings

Auto-Update

• Update notification — A slide-in notification appears when a new version is available, with an "Update Now" button

Multi-Monitor

• Cursor-aware positioning — The floating icon now appears on whichever monitor your cursor is on

New Models

• GPT-5.4 — Added as the new flagship OpenAI model

• Qwen 3.5 — New local models added; removed sub-1B models

Bug Fixes

• Fixed paused media resuming unexpectedly on Windows when no active playback sessions exist

• Fixed Windows hotkey listener state corruption when switching hotkeys

• Fixed silence detection rejecting valid speech (lowered threshold)

• Fixed Windows paste not working in Windows Terminal (scan codes now included)

• Fixed API keys saved in Settings being overridden by shell environment variables

• Fixed local LLM models being deleted during Windows app updates

• Fixed hotkey tooltip display for multi-key combos on macOS

• Fixed RPM install conflicts with other Electron apps on Linux

• Added Tailscale VPN (CGNAT range) support for self-hosted setups

Other

• Panel start position now persists across app restarts

• Cross-window settings sync (hotkey changes apply instantly)

• Account deletion flow with cloud data cleanup

• Agent mode window renamed to "Agent Chat"

Clearer Permissions

• "Screen Recording" → "System Audio" — all permission prompts now clearly state we capture other participants' audio, not your screen

• Electron 39 — eliminates the purple "screen recording" indicator, the "Your screen is being observed" lock screen message, and the misleading permission prompt on macOS 14.2+

Better Soft Voice Recognition

• Auto Gain Control now enabled for dictation, automatically boosting quiet speech

• Lower VAD sensitivity so soft-spoken audio is no longer missed

• Less clipping at the start of speech — increased padding so quiet beginnings aren't cut off

Linux

• Hyprland Wayland support — native global shortcuts using hyprctl keybindings + D-Bus

Bug Fixes

• Fixed wl-copy failing silently on Wayland due to a too-short timeout

• Fixed media staying paused after recording silence with "Pause media on dictation" enabled

System Audio Capture

Smarter Meeting Detection

Bug Fixes

• Fixed hotkey reliability on Windows 11 (modifier-only shortcuts)

• Fixed macOS Globe key failing silently on fresh installs — now prompts for accessibility permission

• Fixed initial audio being dropped during realtime streaming

• Fixed custom dictionary errors on large dictionaries

• Fixed Parakeet model extraction on Windows 10

• Even faster transcription on OpenWhispr Pro + Websockets for gpt4o

• Bug fixes

• Agent mode! Separate hotkey to spin up a chat, now you don't have to open a claude/chatgpt window. Check it out in settings.

• Link your Google calendar and record meetings without a bot

• Search your transcripts and notes

• Brought back the ability to clear all transcripts

• Choose voice recorder start position

• Toggle to minimize the dashboard on startup

• Toggle to pause music etc. when dictating

• A bunch of other bug fixes and performance improvements

• Added Google Gemma local LLMs

• Bug fixes and performance improvements

• Word edit detection + auto add to dictionary

• Bug fixes and performance improvements

• Bug fixes and performance improvements

• CUDA support for whisper-server (local whisper models)

• GPU support for llama.cpp (Local LLMs)

• Other small bug fixes and performance improvements

• Create and manage notes

• Upload audio files and automatically transcribe them into notes

• AI actions in notes

• Referral scheme (give 1 month get 1 month)

• Improved dark mode

• Japanese as an interface language

• Paste improvements on Windows

• Better support for Simplified vs Traditional Chinese on OpenWhispr Pro

• Reliability improvements for OpenWhispr Pro

• Better auto-paste support for Linux users

• Better language support for OpenWhispr Pro

• Internationlization ( EN, PT, IT, ZT, DE, ES, RU, ZH)

• Bug fixes & performance improvements

• Bug fixes and performance improvements

• Bug fixes and performance improvements

• Faster startup — Lazy-loaded heavy components and deferred non-critical init so you get to dictating quicker

• Redesigned toasts — Dark HUD-style notifications that actually look like they belong here

• New loading screen — Branded splash with a proper spinner instead of staring at nothing

• Dictation sound effects — Optional start/stop audio cues (toggle in settings) with warm, soft tones

• Auto-hide floating icon — The dictation icon can now get out of your way

• Smoother scrolling — Thinner, floating scrollbar thumbs that fade in when needed

• Dark mode fixes — Dropdowns and toasts now respect your theme properly

• Auth-aware routing — If your session expires, you'll see a sign-in screen instead of a void

• Globe key persistence — Your hotkey config actually sticks now

• Discord support — Find us in the support menu

• Bug fixes

Features

• Mistral Voxtral Transcription — Transcribe with Mistral's Voxtral Mini model, including custom dictionary support for better accuracy with names and technical terms

• Windows Modifier-Only Hotkeys — Use standalone modifier keys (like Ctrl or Alt) as your dictation hotkey on Windows

• Custom System Prompts — Your custom system prompt now works with local and Anthropic BYOK reasoning

• Audio Cues — Dictation start/stop sounds are more audible

Bug Fixes

• Fixed clipboard paste (Ctrl+V) failing after transcription on Linux Wayland

• Fixed transparent window flickering on Wayland and X11 compositors

• Fixed app hanging during update installation

• Fixed language selector dropdown positioning in settings

• Add language selector to settings

• OpenWhispr Pro!

• Improved language support

• Dark mode!

• Bug fixes