Whisper Notes - Speech to Text

- Auto-skip silent segments in recordings for more stable transcription of long audio files. - Enhanced data storage stability. If data is lost, you can recover unsaved data in Settings - Data Recovery.

- Transcription results now support streaming output. - Optimized search performance for improved product experience.

Refining product UI and UX—I love the polishing process.

- Improved recording stability. - You can now edit/modify transcription results more easily. - Fixes: Fixed recording issues with AirPods/Bluetooth connections during transcription. Fixed some localization problems.

Better recording stability across various scenarios.

New Features - Language Pinning: Long-press to pin your most-used languages to the top of the list. - Smart Folders: Organize recordings by category — All, Recordings, Imports, and Favorites. - Audio Waveform: Visualize and seek through audio in the detailed transcript view. - Live Progress & ETA: See real transcription progress with estimated time remaining. - Export TXT: Export transcripts as plain text files with timestamps. Improvements & Fixes - Fixed audio routing issues when switching between speaker and headphones. - Fixed recordings being interrupted by alarms or phone calls. - Fixed imported audio files being incorrectly categorized as recordings. - Fixed UI flickering when returning from Live Activity. - Improved VAD (Voice Activity Detection) to reduce hallucinations during silence. - Redesigned Settings page with clearer organization. - Added Reddit community link for support and feedback.

New Features - Dark Mode App Icon: Switch between light and dark app icons in Settings. - Floating Recording Window: Swipe right during recording to minimize to a compact floating widget. - Real-time Audio Waveform: See live audio levels while recording. - Search Category Filter: Filter search results by category — All, Recordings, Imports, or Favorites. Improvements & Fixes - Fixed recording view not responding properly to Lock Screen widget. - Fixed "Instant Record on Open" triggering on brief app switches. - Improved haptic feedback and audio waveform visualization.

New Features - Share Extension: Import audio and video files directly from the system Share Sheet. - Tap-to-Seek: Tap any sentence in a transcript to jump to that exact moment in the audio. - iOS 26 Liquid Glass: Adapted to Apple's new translucent design language. Improvements & Fixes - Background music now ducks during playback and pauses during recording, instead of stopping entirely. - Fixed tap-to-seek highlight not aligning with the correct audio position. - Fixed recording view not dismissing when stopping via Live Activity. - General performance improvements and bug fixes.

New Features - Parakeet v3 as the default recognition engine — 10× faster, with much more accurate transcriptions and fewer hallucinated words. - Switch between Whisper and Parakeet models in Settings. Improvements & Fixes - Exported audio/video files now preserve original filenames. - Fixed audio extraction failure for imported videos. - Recordings are now safely saved if the app exits unexpectedly.

New Features - Live audio waveform with a health indicator that warns when no audio is detected. - Auto Copy now lets you choose: Off, Text Only, or Text + Date & Time. Improvements - Long transcriptions are now automatically split into paragraphs for easier reading. - Improved stability and reliability when downloading models.

Improve recording stability and crash recovery.

New Features - Auto-Stop Timer — Set a recording time limit (30 min to 6 hours). Recording stops and saves automatically when time is up. - Siri & Action Button — Start recording hands-free with "Hey Siri, start Whisper" or assign it to your Action Button. - Custom Dictionary — Add proper nouns and specialized terms to improve transcription accuracy. Improvements - Recording now captures a little extra at the end to avoid cutting off the last word. - Fixed a crash that could occur when running in the background. - Settings page reorganized for easier navigation.

Improvements - Transcription is now noticeably faster after the first use — no more repeated model loading. - Recording starts quicker — less delay between tapping the button and hearing your voice captured. - Share cards now let you hide the app watermark. - Fixed several crashes during recording haptics, model extraction on older devices, and background transcription.

New Feature - Filler Word Removal — Automatically clean up "um," "uh," and other filler words from your transcriptions. Improvements - Longer recordings are now significantly more accurate — reduced repetition and language-switching errors. - Bluetooth recording quality improved — noticeably clearer audio with AirPods and other Bluetooth devices. - Lock Screen widget is now more responsive — no more stuck "Processing" state, and recording starts faster. - Auto Copy now works reliably when switching to other apps. - Editing a transcription no longer overwrites the original text. - Fixed crashes that could occur during transcription.

Improved recording stability — fixed crashes during transcription and reduced audio start delay when using Action Button.

Improvements - More reliable recording when switching between speakers, headphones, or Bluetooth devices. - Fixed clipboard not working reliably after transcription. - Imported audio files are now verified before cleanup, preventing accidental data loss. - Custom Dictionary now handles large word lists without freezing. - Spoken punctuation ("new line", "period") now works correctly during editing.

New - New speech model option with improved accuracy for Chinese, Japanese, Korean, and Cantonese. Improvements - Each model now remembers your language preference separately. - Fixed Dynamic Island staying active after recording ends. - Improved transcription display and paragraph spacing.

New Features - Lock Screen & Control Center playback — play, pause, and skip through your recordings without opening the app. - New storage management — see what's using space (audio, transcripts, models, cache), empty the recycle bin, and delete downloaded models you no longer need. - Search now sorts by newest, oldest, or relevance, with clearer match highlighting. Performance & Recording - Faster app launch, and large histories now load smoothly as you scroll. - Fewer freezes opening long transcripts, and better memory handling on older iPhones during long transcriptions. - More stable recording when switching between the built-in mic, AirPods, and other Bluetooth devices. - Simpler, more natural default recording settings. Transcription & Playback - Improved automatic paragraph breaks, including for Chinese/Japanese/Korean and long text without punctuation. - More accurate spoken punctuation, and smarter "No text detected" handling with automatic retry. - More consistent timestamps for playback highlighting and SRT export. - Tap any line in a transcript to jump to that moment in the audio. Stability & Fixes - Added speech-model crash recovery with automatic fallback for more reliable transcription across devices. - Safer storage cleanup that won't mistakenly remove your recordings; fixed duplicate models and stuck processing items. - Reduced the chance of losing a message when importing or transcribing right after launch. - Fixed the "Copied" badge sometimes staying on screen, plus several playback display glitches. Accessibility - Added Magic Tap to start/stop recording, plus VoiceOver improvements.

This update focuses on speed, recording reliability, and transcription quality. Performance - Faster app launch — transcription models load about twice as fast, and your notes appear instantly even with thousands of recordings. - Long recordings with pauses transcribe much faster — silent stretches are skipped instead of processed. - Recording starts at full speed the moment you open the app. - Switching transcription models is now smooth, with no freezes. Recording reliability - More reliable recording with AirPods and Bluetooth — less audio lost when you connect or remove a headset mid-recording, and faster recovery if it stops sending audio. - Fixed rare issues with rapid stop-and-restart recording and recordings interrupted by phone calls. Transcription quality - Live transcript lines now always appear in the correct order during long transcriptions. - More accurate timestamps in recordings with long silences. Stability - Fixed the most common crash — transcribing while the app is in the background — plus several rarer reliability issues.

New Features - Speaker Labels — Whisper Notes now labels who said what, automatically splitting any transcript into color-coded speakers you can rename, even across short interjections and quick back-and-forth. Runs 100% on your device, and speaker labels are included in copied text and exported SRT / VTT subtitles. - Folders — organize your recordings into folders you create. Anything you record or import inside a folder is filed there automatically, with batch move and per-folder search. Design - A warmer, easier-on-the-eyes look — soft paper tones instead of stark white, and gentler text in dark mode. - More refined haptic feedback throughout, and cleaner Share and Export menus. Reliability - If storage runs low while recording, you're now warned instead of ending up with a silent, empty file.