Updated on Jun 2, 2026
Improved on-device transcription performance. Audio processing now starts faster, model loading is reused more efficiently, and shared audio files should reach displayed text with lower latency while keeping the same app behavior.
