Improvements to the Duo experience

• Added support for iPhone Duo, with layouts that adapt when you open or fold the device. • Browse transcription and dictation history in a two-column grid on larger displays. • Improved the transcript information panel so opening and closing it keeps your place.

■ New • New AI models: GPT-6 Sol and GPT-6 Luna, Gemini 3.7 Flash and Gemini 3.8 Flash, and Claude Opus 5.5. • SpaceXAI (Grok) is now available as an AI service, with Grok 4.7 as the default model. ■ Improvements • The AI Services settings screens are redesigned, with a service icon for each provider. You can also set up more than one configuration for the same service. • New AI services default to newer models: GPT-6 Luna for OpenAI, Gemini 3.8 Flash for Gemini, Grok 4.7 for SpaceXAI. • ElevenLabs cloud transcription now takes files up to 3 GB (was 1 GB) and up to 10 hours long (was 4.5 hours, or 2 hours with speaker detection). • Settings now explains that your OpenAI key can also be used for cloud transcription. • The ElevenLabs API key now has to be saved explicitly, so it can't be half-entered by accident. ■ Bugfixes • The Stop button on the dictation Live Activity now works even after the app has been closed.

■ Bugfixes • Fixed Whisper Transcription actions not appearing in Shortcuts and Siri AI. ■ Improvements • Added the latest OpenAI and Anthropic models, including GPT-5.6 and Claude Fable 5.1. • Added missing translations throughout the app. • Refined the keyboard setup screen.

■ New • Siri & Shortcuts: Find and open transcriptions, or use transcript text in your shortcuts. • Apple Intelligence: Find and search through your transcripts by talking to Siri AI in natural language and give Siri knowledge about your transcripts. • Spotlight: Search your transcripts directly from Spotlight, with an option to disable indexing in Settings. • Google Gemini: Use your own API key for summaries, smart titles, and transcript chat. ■ Improvements • Playback: Refreshed playback and chat controls on iOS 26, with better support for larger text sizes. • Search: Find transcriptions by their original filenames, AI-generated titles, and summaries. • Languages: The language picker now shows only languages supported by your selected transcription model. ■ Bugfixes • Fixed: Downloaded on-device models could unexpectedly require an internet connection to load. • Fixed: Retranscribe could do nothing after importing a file. • Fixed: Failed or cancelled transcriptions could hide or discard their recordings. • Fixed: Deleted recordings could remain in storage or appear in history. • Fixed: Opening API key settings could unexpectedly switch your preferred AI service to Anthropic. • Fixed: Subscribe to Assistant and Other options could fail to open the app from the dictation keyboard.

■ New • Import from URL: paste a link to an audio or video file, and transcribe it straight from the home screen • Added Claude Opus 5 to the Anthropic model picker for summaries, titles, and transcript chat ■ Improvements • Redesigned Settings with new icons • Assistant transcription now runs on Mistral's Voxtral model; larger files are still handled by ElevenLabs Scribe • Clearer errors when an on-device model download fails because the connection is blocked or unstable • Updated German, Dutch, and French translations ■ Bugfixes • Fixed the detected language not showing on transcripts made with the Assistant • Fixed some audio files being rejected by the Assistant because the upload did not include the original filename • Fixed the "file too large" message from the Assistant not being translated

■ New • Added export previews, with support for VTT and Markdown subtitles and DOTE files • Added Anthropic API key support and model selection for summaries, titles, and transcript chat • Added an OpenAI model picker, including support for custom models • Transcripts can now be renamed manually ■ Improvements • Redesigned transcript chat with clearer loading and empty states, a better input bar, and easier response copying • Improved the transcript detail layout on iPad, including a wider playback bar and a collapsible inspector • Improved search result reliability and transcription retry actions • Improved the confirmation shown when deleting all dictations • Updated translations throughout the iPhone and iPad apps ■ Bugfixes • Fixed the selected AI service sometimes changing when it was temporarily unavailable • Fixed opening Custom API Keys unexpectedly switching the selected AI service to OpenAI • Fixed unclear errors when an on-device model download became corrupted • Fixed sharing a transcript chat response sometimes sharing the prompt instead of the generated answer

■ Improvements • A cleaner, more consistent look across the app, including a redesigned on-device model download screen. • Transcribe audio-only MP4 files. • Smoother search on the home screen. • Sharing a video now shows "Share Video" with a video icon, audio files show audio. • More room on iPad: the home screen shows more of your history with longer, multi-line previews that fill the screen. • Quick paste buttons on the OpenAI and ElevenLabs API key fields. • VoiceOver and accessibility improvements throughout - progress indicators and the playback scrubber now announce what's happening (for example "1:23 of 4:56"). ■ Bugfixes • Fixed short audio clips sometimes producing an empty transcription. • Fixed the Share Sheet prompt on the home screen dismissing when you didn't mean it to.

■ New • Local transcription models now run on latest iPad models ■ Improved • Assistant can now handle files twice as long

■ Improvements • Improved stability when using local models

Due to system changes the dictation feature can no longer start automatically. You will have to move back to the previous app manually by swiping on the bottom of your screen. Sorry about that! ■ Improvements • Dictating with local models is more reliable ■ Bugfixes • Fixed an issue where dictation could not be activated. • Fixed an issue where using your own API could require a pro subscription in some scenarios

- Fixes an error alert that could occur when downloading the local model

■ Improved • You can toggle off the summary view on the homescreen • When adding an API key, it’s automatically enabled ■ Bugfixes • Fixed an issue where if you didn’t have an AI service setup, you could get an error alert when creating a summary

■ Improved • Small tweaks and design improvements on iPadOS

■ New • Implemented a new Liquid Glass design • You can now browse your old transcriptions and dictations • Added support for Apple Foundation Models to be used for AI features • Added support for the new on device transcription engine • A short summary is now autogenerated if you have an AI service setup • You can adjust the prompt that is used for the long summary • Added initial support for Shortcuts. We’ll add more soon. • Recordings are automatically saved, even if they fail for some reason. ■ Improved • The app is now much faster to use • Added support for more apps for the dictation keyboard ■ Fixed • Fixed an issue where a recording would not be saved

■ Fixed • Fixed an issue where the local model could not be downloaded

■ Improvements • Added support for more apps for the keyboard feature ■ Fixed • Fixed a crash that could happen when running the app from a locked phone state

■ Improvements • Added a check for minimum available storage space before you can download a local model. This could lead to crashes when downloading without having enough space. • Cleaned up the dictation history detail view • Improved dictation support for Microsoft Teams, Outlook, Claude, Drafts 5 and more. ■ Fixed • Disabled swiping in the keyboard setup view

■ New • This update introduces a new dictation keyboard you can use in any app on your iOS device! Simply select it from the Globe icon after setting it up and you can talk much faster than you can type all across the system. Let us know how it works for you. We’ll be improving it a lot over the next few weeks based on your feedback!

Fixed an issue where the summary could be shown in the incorrect order

■ Fixed • It was possible for transcription segments to sometimes be displayed in the wrong order when transcribed on-device.

■ New • Added support for videos • Improved performance for local models ■ Improvements • Improved support for files that were stored in iCloud when shared with the app • Audio playback improvements

■ New • Your transcriptions and recordings are now saved in the new history view. Quickly revisit previous transcriptions without having to transcribe them again ■ Improvements • Improved design stuff all around the app, making it more clean and simple • If you’re playing music it will no longer be paused when you get to the transcription view • Added a “Are you sure you want to quit alert” to prevent data loss • Clearer error alerts with simpler descriptions ■ Bugfixes • Voice Memos audio files no longer show the WhatsApp icon • Fixed an issue where the summary could be presented out of order

■ Improvements • Improved the German, Dutch & French localisation • More accurately show the origin app of the file you share to the app

■ Improvements • Improved the navigation flow of the app while transcribing files from other apps ■ Bugfixes • Due to technical changes the file size limit for cloud transcription is 50MB right now. We’re looking into more ways to increase this.