A new speech engine and a major new capability: Hedy can now identify who is speaking and show it in the transcript.

Transcripts now show who said what. Each speaker gets a unique color and label, and the transcript is displayed turn by turn — grouping what each person said together. This makes it much easier to follow multi-person conversations.

The speaker identification is powered by a new speech engine based on NVIDIA's Parakeet TDT v3 model, which also offers improved transcription accuracy compared to Whisper.

Parakeet also works when importing audio files, with full speaker diarization applied to the imported audio.

Currently available on macOS with Apple Silicon only. You can switch back to Whisper at any time in Settings.

Your session audio recordings now sync automatically via iCloud across your Apple devices.

Windows users can now import video files directly for transcription.

Supported formats include MP4, AVI, MOV, and 3GP. The audio track is automatically extracted from the video and processed through the speech engine. The original video is not stored — only the extracted audio.

Powered by the native Windows Media Foundation framework. If Media Foundation is not available on your system (e.g., Windows N editions), Hedy will show a notice with instructions to install the Media Feature Pack.

Detailed notes now include a "People & Context" section that captures the personal and relationship-building details shared during a conversation:

This section only appears when personal details were actually shared, so purely transactional meetings won't have it. Sensitive information (health, finances) is never included.

Especially helpful when reviewing notes from meetings weeks or months later, reminding you of the personal context around professional conversations.

You can now create a new session context directly from the context dropdown when starting a session.

Select "Create New Context" from the dropdown, enter a title and description in the editor that appears, and save. The new context is automatically selected for your session. No need to navigate to Settings first.

Free users can create one context. Pro users can create unlimited contexts.

The session detail view on desktop now offers three preset layout modes, accessible via a new button in the chat panel header:

Transitions between modes are smoothly animated. Your selected mode is remembered across sessions.

The export and import experience has been completely redesigned.

JSON exports now capture everything in your Hedy account:

You can export as a single JSON file or as a ZIP with separate files per session.

This makes it easy to migrate your complete data from one Hedy account to another. Export everything from your current account, sign into your new account, and import the file. This is especially useful if you need to switch to a new EU-region account for data residency.

Real-time coaching suggestions during a session no longer bleed into your summaries, detailed notes, or chat responses. Previously, suggestion content could make AI-generated notes feel repetitive or off-topic. Now, suggestions are filtered out before generating any downstream content, so your notes and summaries stay focused on the actual conversation.

The suggestions themselves still work as before during the session. Your notification preference is also now correctly preserved when toggling the feature on and off.

Editing a session's date or time is now faster and more intuitive.

No more clunky picker dialogs on desktop.

Topics, highlights, and chat history now load significantly faster using real-time delta sync and cache-first reads, especially when offline or on slow connections.