treatAllFilesAsSingleChannel
Set this option to "true" to merge all the channels of a multi-channel file to a single file before processing. Normally each channel of a multi-channel file is processed by ASR as if it were a separate single channel file, each channel is then added to the item as a separate recording with a single speaker per recording with the label Channel 1 .. Channel 2 etc.
NOTE that when a multi-channel file is processed as separate channels the segmentation data has a single segment for each recording that is as long as the recording so you cannot determine the change of speakers using the segmentation data alone you must order the srt words by timestamp and then detect change of speaker yourself.
File types
There are 3 sorts of files:
Multi-channel: Multiple speakers separated into multiple channels. If you open the file in Audacity you will see a different waveform in each channel. The settings for this are treatAllFilesAsSingleChannel=false, diarization.enabled=false
Stereo: One or more speakers, but with the same audio in each channel. If you open the file in Audacity you will see the same or similar waveform in each channel. The settings for this are treatAllFilesAsSingleChannel=true, diarization.enabled=true
Mono: Only one channel. If you open the file in Audacity you will see only one channel. The settings for this are treatAllFilesAsSingleChannel=true, diarization.enabled=true