It is true that there is a phased delivery of Transcription based on Edge deployment type.
Our upcoming beta and first release will provide native transcription for deployments where Edge is in the Cloud. That means either BYOCC or PCV/GCV. For these deployment types the audio stream is sent in real time to the Transcription Service to transcribe in real time and have the transcript available immediately after the call completes.
Subsequently, we will add support for BYOCP. For these deployment types, we wait for the recording to be uploaded in to the Cloud and then perform transcription against the recorded audio. In these deployment types there are often bandwidth constraints that limit the ability to stream in real time. As a result, the architecture is slightly different. The precise timeline hasn't been determined yet, but we are targeting Q4 for that deployment type.
There is an existing idea in the Ideas Lab for this, it would be helpful to vote on it to help prioritize it further:
https://purecloud.ideas.aha.io/ideas/CLWFO-I-276------------------------------
Rakesh Tailor
Director, Product Management - Workforce Engagement
Genesys Cloud
------------------------------