Dataset extracted from audio recordings from predator presentations (first 50 calls)

Dataset used to run the code (also in the FigShare depository), extracted from 58 audio recordings of the vocal reaction of titi monkeys when presented to three stuffed predators (tayra, oncilla and caracara).

Coding methodology: We've used the vocal repertoire described by Cäsar et al. (2012), which includes two alarm calls: A- and B-calls. Audios were listened with Praat, and the first 50 calls of each sequence were labelled as A-calls, B-calls or "other" (undetermined call or other call type). The beginning and ending time of each call was also labellised.

You can find the recordings collected by Berthet et al. (2019) in the following link : https://figshare.com/articles/dataset/Audio_recordings_of_predator_presentations/7707119. Recordings collected by Cäsar must be requested directly to the author.