normalize the gain by transcription identified speakers
I would like each speaker voice group to be used for audio gain normalization.
In the Zoom meeting I edited there were multiple participants whose microphones were all at different levels. After using the Speech to Text in the beta it identified each speaker voice by number.
I had to visually look at the audio waveform and search for differences to identify each speaker. Then I cut with the Razer Tool around each voice and normalized the gain after selecting each segment.
But since the speech to text recognition identifies each source, perhaps those groups of speakers could also be used to create audio groups for normalization.