Improved multiple birdsong tracking with distribution derivative method and Markov renewal process clustering
This work addresses the challenge of automatic bird sound recognition, which is important for ecological monitoring and bioacoustics, but it appears to be incremental as it builds on existing methods with specific improvements.
The paper tackled the problem of segregating multiple simultaneous bird sounds in audio mixtures by using an improved spectrogram representation based on the distribution derivative method, which enhanced the performance of a segregation algorithm using a Markov renewal process model for tracking vocalization patterns.
Segregating an audio mixture containing multiple simultaneous bird sounds is a challenging task. However, birdsong often contains rapid pitch modulations, and these modulations carry information which may be of use in automatic recognition. In this paper we demonstrate that an improved spectrogram representation, based on the distribution derivative method, leads to improved performance of a segregation algorithm which uses a Markov renewal process model to track vocalisation patterns consisting of singing and silences.