Retrieve the speech recognition transcription for a single call, by call ID.

When to use. Fetch a finished transcription on demand. If the call is still being transcribed, the response indicates it is in progress; if the call was never sent to speech recognition, the request returns a not-found error.

Parameters. call_id (path, required): the numeric call ID.

Returns. A data object with the transcription and its analysis:

• plain_body (string): full transcription text
• phrases (array): detected phrases, each with phrase, start_at, end_at and channel
• summary (string): AI-generated summary of the transcript
• sentiment (float): call rank from 0 to 1, assessing the quality and value of the call
• language (string), created_at (string), remote_identifier (string)
• metadata.channel_id (string): the speech channel the call belongs to
• status (string) and message (string)

Note. Only the token used to submit a call can retrieve that call's transcription.