How to apply this guide
Choose a recording channel based on the actual mic path and check that important speech survives export.
Start with
Identify where the microphone appears in the recording. If only one channel contains speech, listen to it by itself rather than assuming a stereo mix is better. If separate voices matter, archive the original channels.
Check before you finish
Store source audio according to your policy and label the mono file as a processing copy. Use labels in notes if speaker attribution matters later.
Evidence to keep: Keep the target field open and verify that the complete reviewed result reached the intended destination. This page’s key limit is: Choose a recording channel based on the actual mic path and check that important speech survives export.
Start here
A single-speaker recording often needs only one usable speech channel. Stereo can be useful when different inputs must remain separate. Decide what to preserve before making an import copy.
Check the capture setup
Identify where the microphone appears in the recording. If only one channel contains speech, listen to it by itself rather than assuming a stereo mix is better. If separate voices matter, archive the original channels.
Make a clean comparison
Export a short mono copy from the speech channel using a compatible format. Run it and the original through the same language and model. Check quiet words and the closing phrase.
Keep the original available
Store source audio according to your policy and label the mono file as a processing copy. Use labels in notes if speaker attribution matters later.




