Improvements

• Improved accuracy for transcription of medical terminology for English (Enhanced operating point only)

• New models for Vietnamese and Portuguese with improved transcription accuracy (Standard operating point only)

• New models for Standard Operating Point, with improvements in discriminating between speech and non-speech audio

  • Arabic (ar), Mandarin (cmn), Danish (da), German (de), Spanish (es), French (fr), Hindi (hi), Hungarian (hu), Indonesian (id), Italian (it), Japanese (ja), Korean (ko), Dutch (nl), Norwegian (no), Portuguese (pt), Russian (ru), Swedish (sv), and Cantonese (yue)

• End of Utterance and Audio Filtering configuration settings can now be updated during an active session

Fixes

• New models for English with improved accuracy in transcribing initialisms

  • Resolves tickets: 27748

• Custom Dictionary errors are now handled gracefully, ensuring the session continues without interruption