AUTOMATIC TYPE CLASSIFICATION AND SPEAKER IDENTIFICATION OF AFRICAN ELEPHANT VOCALIZATIONS
This paper presents systems for automatically classifying elephant vocalizations by type and for identifying the speaker of a given vocalization. The method applies techniques from the speech processing field, with modifications, to elephant vocalizations. The features used for classification are 12 Mel-Frequency Cepstral Coefficients computed using a chirp Z-transform to interpolate among the lower frequencies. A Hidden Markov Model is trained for each type of vocalization and vocalizations are classified using leave-one-out verification. Using this system, initial classification accuracies of 77.0% for type classification and 72.2% for speaker identification resulted. These systems represent the initial stages of a universal analysis...
- Nenhuma restrição
- Outra biodiversidade terrestre
- Artigo com revisão pelos pares e teses universitárias
- Identificação, Monitoria, Indicadores e Avaliações
- Todas Áreas de Conservação de Moçambique