Pro users now get automatic AI suggestions for up to 120 minutes per session, doubled from the previous 60-minute limit.

Hedy's cloud-based speech-to-text has been upgraded to a newer, more capable engine for improved transcription accuracy and reliability.

Starting a session on macOS is now smoother. We fixed an issue where audio capture initialization could briefly freeze the app on session start.

When you switch your audio output device on macOS (for example, connecting AirPods), Hedy now automatically restarts the microphone alongside the system audio to keep your recording uninterrupted.

Resolved a race condition that could cause audio playback to be unavailable on some sessions. Recorded audio now loads reliably every time.

Changing your email address is now more reliable. We fixed an issue where the process could fail partway through, leaving your account in an inconsistent state.

You can now choose to store your data in the European Union. When you select EU data residency, all your sessions, transcripts, and topics are stored and processed exclusively in EU data centers.

Once set to EU, your session data never touches US servers.

This feature is only available to new users while creating a new account. We are evaluating migration options for existing users.

New users are taken through a guided onboarding experience to help create their first personalized session context through a series of interactive questions about their role, goals, and challenges. Hedy uses those answers to generate a tailored AI context that makes coaching more relevant to their specific situation.

Windows now uses your discrete GPU (NVIDIA, AMD) for Whisper speech recognition more reliably, delivering significantly faster transcription performance.

Session types are now grouped by category, making it easier to find the right coaching mode for your situation. Improved descriptions help you pick the best session type for each conversation.