Acoustic-Based Applications for Vertebrate Vocalization

The vocal sound production of vertebrates is considered the most definite form of communication. Vocal communication, which includes ‘calls’ and ‘songs’, is mainly used for various purposes such as location identification, food search, obstacle detection, warning calls and courtship. Vocalisation analysis of vertebrates contains many challenges, and some of these challenges are still not addressed. The data recording, processing and best acoustic feature selection can be categorised as the most difficult challenges faced in past and recent studies. However, there is a high possibility to resolve these challenges by integrating the available technologies. Thus, this integration can improve the quality of data recording instruments and process the recorded data accurately to find novel acoustic features of each vertebrae category. Almost all the studies still depend on the regional and species-based acoustic pre-recorded data commonly used in passive acoustic monitoring applications. A vast spectrum of work needs to be addressed in active acoustic monitoring-based applications. This chapter reveals the vertebrate vocalisation-based applications that must need more improvements in future.

The acoustic features are important to be extracted from vertebrate vocal signals to achieve the most accurate results from the related applications applied in various fields such as medical, industrial, biological and many more. Specifically, homomorphic signal processing features such as MFCC, LPCC, GFCC, GTCC and PNCC have taken significant popularity in deriving acoustic features that can be applied in numerous signal processing applications, including speech recognition and image enhancement. This chapter elaborates the extraction process of potential acoustic features in the time domain, frequency domain and cepstral domain. Then it will further discuss the feature engineering and selection techniques that can be applied to the acoustic features.

The technological developments have opened more multidisciplinary opportunities worldwide, especially in the medical and engineering fields. Apart from many other opportunities, audio signal processing takes a significant role because it can provide a broader range of applications with the integration of trending technologies such as artificial intelligence (AI), augmented reality and virtual reality (ARVR), cognitive computing, internet of things (IoT), intelligent apps (I-Apps) and robotic process automation (RPA). However, these technologies need to be applied in vertebrate vocalisation-based applications, including monitoring species biodiversity, conserving endangered species, tracking environmental pollution, precision livestock farming, frightening tools and biology-based studies. This chapter showcases some vertebrate vocalisation related applications that can be newly invented via trending technologies. In addition, this chapter provides some suggestions to improve a few existing vertebrate vocalisation related applications. Thus, these applications may cooperate to various fields of studies to get extreme benefits, especially in medical, industrial, agricultural, scientific research works and many more to improve their findings and innovations.