Guinea baboon vocalisations dataset automatically extracted from natural audio recordings

The data collection process consisted of continuously recording an enclosure of Guinea baboons for one month. They live in semi-liberty at the CNRS primatology centre in Rousset. A microphone was placed nearby to record the sounds produced by the group without the presence of a human. A neural network was used on these large and noisy audio recordings, following the method of Bonafos et al. (2023). The neural network automatically detected segments in the data stream when the baboons produced vocalisations. The dataset consists of one second to several minute wav files of automatically detected segments of vocalisations produced in semi-natural environments. The dataset thus provides a new set of baboon vocalisations produced at all times of the day, without the presence of a recording human. It can be used to learn more about the vocal productions of non-human primates, their distribution over the day, their frequency, and their heterogeneity. In addition to the analysis of animal communication, the dataset can also be used as a learning base for sound recognition models.