[FEATURE REQUEST] Add optional speaker information with whisper transcription

[Jiltseb]

Feature Summary

• This feature will add extra processing to take the output from diarization and whisper deployments and combine them together to form transcription with speaker information.

Justification/Rationale

• Diarized transcription adds additional speaker information and this helps the LLM to infer speaker-related information and associate it with the transcription

Proposed Implementation (if any)

• Add speaker processor that has speaker-related utility functions. One utility function should handle post-processing for diarized transcription. Since this is an app with multiple deployments, a notebook needs to be created to show how it works.