How would you analyze the sound track in this video to show the presence or absence of acoustic interactions between the cello emissions and the birds?
$\begingroup$
$\endgroup$
2
-
2$\begingroup$ One can imagine that Yo Yo Ma changes its way of playing while listening to the birds; is there any background to hypothesize that the birds would change their vocalizations in response to such cello sounds? $\endgroup$– NoilCommented Jul 23, 2022 at 21:19
-
$\begingroup$ Good point Noil. Maybe this possibility makes it more difficult and interesting. Also maybe we can consider this as any two sets of individuals possibly interacting acoustically. And to make it harder, which bird species members appear to be interacting and which ones most probably are not. $\endgroup$– sm1Commented Jul 23, 2022 at 21:55
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
Direct answer: I would use the transfer entropy analysis, as used by Reiji Suzuki and colleagues (Suzuki et al 2017). I also developed a method for analysing interactions (2016) but my method is for short-duration events (calls) and is not appropriate here. Transfer entropy should be appropriate for detecting influences between parallel ongoing streams of sound.
You need to choose an appropriate feature representation of the audio. Here, I would simply use the pitch track.
Additional notes:
I wouldn't expect to discover much interaction. The most prominent bird sound seems to be a nightingale, a European bird which I wouldn't expect to encounter in the habitat nor time of day shown in this video! The nightingale is presumably prerecorded and mixed in later (just like the backup cellos), which for me is a bit of a disappointment.
You can read David Rothenberg's "Why Birds Sing" for some extended discussion of whether or not we would expect birds to interact with musicians, written by someone with experience of playing the clarinet "with" birds.
answered Aug 17, 2022 at 12:35