How do I retrive the speech text?

Running starter program, it detects the speakers. How do I get the speech?

My usecase is I like to output a transcript in following form

Start_time, End_Time, SpeechText, Speaker_id

Start with the starter program referenced in the issue and inspect how it exposes speaker-detection results and whether speech text is available. Done means the repository documents or demonstrates obtaining Start_time, End_time, SpeechText, and Speaker_id, or clearly explains if speech transcription is outside the program's scope.

Written by the indexing model from the issue text.