This paper presents a method for automatic classification of birds into different species based on the audio recordings of their sounds. Each individual syllable segmented from continuous recordings is regarded as the basic recognition unit. To represent the temporal variations as well as sharp transitions within a syllable, a feature set derived from static and dynamic two-dimensional Mel-frequency cepstral coefficients are calculated for the classification of each syllable. Since a bird might generate several types of sounds with variant characteristics, a number of representative prototype vectors are used to model different syllables of identical bird species. For each bird species, a model selection method is developed to determine the optimal mode between Gaussian mixture models (GMM) and vector quantization (VQ) when the amount of training data is different for each species. In addition, a component number selection algorithm is employed to find the most appropriate number of components of GMM or the cluster number of VQ for each species. The mean vectors of GMM or the cluster centroids of VQ will form the prototype vectors of a certain bird species. In the experiments, the best classification accuracy is 84.06% for the classification of 28 bird species.

Automatic Classification of Bird Species From Their Sounds Using Two-Dimensional Cepstral Coefficients

Recognition of Orthoptera species by means of their song is widely used in field work but requires expertise. It is now possible to develop computer-based systems to achieve the same task with a number of advantages including continuous long term unattended operation and automatic species logging. The system described here achieves automated discrimination between different species by utilizing a novel time domain signal coding technique and an artificial neural network. The system has previously been shown to recognize 25 species of British Orthoptera with 99% accuracy for good quality sounds. This paper tests the system on field recordings of four species of grasshopper in northern England in 2002 and shows that it is capable of not only correctly recognizing the target species under a range of acoustic conditions but also of recognizing other sounds such as birds and man-made sounds. Recognition accuracies for the four species of typically 70-100% are obtained for field recordings with varying sound intensities and background signals.

Automated identification of field-recorded songs of four British grasshoppers using bioacoustic signal recognition

Application of time domain signal coding and artificial neural networks to passive acoustical identification of animals

Research into the automated identification of animals by bioacoustics is becoming more widespread mainly due to difficulties in carrying out manual surveys. This paper describes automated recognition of insects (Orthoptera) using time domain signal coding and artificial neural networks. Results of field recordings made in the UK in 2002 are presented which show that it is possible to accurately recognize 4 British Orthoptera species in natural conditions under high levels of interference. Work is under way to increase the number of species recognized.

/pdf/automated-bioacoustic-identification-of-species-20eu9znvh8.pdf

Automated bioacoustic identification of species

Fuzzy systems, neural networks and neuro-fuzzy systems: A vision on their hardware implementation and platforms over two decades

Analysis of speech often begins with study of the vocal tract that created it. Bird vocalizations and human speech are generated by similar processes. This suggests that LPC coefficients extracted from birdsong samples could retain enough information to permit identification of species. In this paper we train a back-propagation neural network to recognize bird songs. We generated test and training data sets using 133 songs from six common bird species. Initially, identification performance was good for some species, and poor for others. We attributed this to a lack of temporal context information in the data. By changing the type of spectral information presented to the network, we were able to improve performance. We conclude that a neural network combined with digital preprocessing can be used to identify a bird by its song.

Birdsong recognition with DSP and neural networks

Today's public switched telephone networks are being upgraded to use digital transmission and switching. The main disadvantage of digital transmission is the higher bandwidth required. To alleviate this problem, much research has been devoted into the compression of digitized voice signals. However, such compression techniques tend to have adverse effects on non-voice signals such as modems and faxes. Only recently has the area of voiceband data signal compression been explored. This paper covers some of the highlights of these publications including DPCM, delta modulation, low delay code excited linear prediction, and vector quantisation.

Survey of voiceband data signal compression

In this paper, we discuss the use of artificial neural learning methods for low bit-rate speech compression, potentially in non-stationary environments. Unsupervised learning algorithms are particularly well-suited for vector quantization (VQ) which is used in many speech compression applications. We discuss two unsupervised learning algorithms: frequency-sensitive competitive learning and Kohonen's self-organizing maps which have both been investigated for learning the codebook vectors in an adaptive vector quantizer. In contrast with earlier work, we have employed these learning rules in VQ of the linear predictive coding (LPC) prediction residual. The performance of these unsupervised learning algorithms in speaker-dependent and speaker-independent speech compression are presented. Our results compare favourably with those of code-excited linear prediction (CELP) requiring reduced computational power with a tolerable reduction in speech quality. We also explore the effects of limited precision on classification and learning in competitive learning algorithms for low power VLSI implementations.

https://ieeexplore.ieee.org/iel3/3680/10821/00494066.pdf

H.C. Card

Papers

Birdsong recognition with DSP and neural networks

Survey of voiceband data signal compression

Neural code-excited linear prediction for low power speech compression