Retirement: Conversation Transcription Multi Channel Diarization

Conversation Transcription Multichannel Audio Diarization will be retired on March 28th, 2025, please transition to Speech to Text Diarization enhanced add-on feature by that date. 

Conversation Transcription Multichannel Audio Diarization was originally designed for multichannel audio from a specific mic array device. To remove the dependency on the special device and simplify the experience, we released Speech to Text Enhanced Add-on feature, Diarization, which provides the speaker diarization capability through mono channel audio. With this new feature, you can get speech to text transcription by speakers (Guest1, Guest2, Guest3, etc.), so that you know which speaker is speaking a particular part in the transcription. We encourage you to transition to Speech to Text Diarization enhanced add-on feature prior to the retirement date to experience the new capabilities of conversation transcription with speaker diarization including using single channel audio instead of multichannel audio, with cheaper price, and simpler implementation, etc.

From now to March 28th, 2025, you can continue to use Conversation Transcription Multichannel Audio Diarization without disruption. On March 28th, 2025, workloads running Conversation Transcription Multichannel Audio Diarization will be deleted and associated application data will be lost.