vocalisation type in BirdNET output

[Dubults]

Currently, the output from BirdNET is the name of the species detected and its confidence level (plus additional technical information depending on the output format selected). However, there are different types of bird calls - songs, alarm calls, flight calls, etc. I assume that all of these are used to train the BirdNET model. If so, BirdNET should be able to provide information on the type of sound in addition to the species name in its output of detections. This would be very important for researchers as it would provide more information on the status of the species at the given location. Also, the confidence of species identification could vary between different types of vocalisations and the researcher could choose different thresholds to use depending on the type of sound, rather than a single threshold for a species. I have checked the Discussions and Issues sections of the BirdNET-Analyzer github and have not seen anything about this. Am I the only one who thinks this would be a much needed feature to consider for future releases of BirdNET? Or have I missed something?

best wishes
Ainars