Concurrent segmentation of multiple similar vocalizations

The method of 2, wherein extracting a respective set of features for each of the plurality of vocalization instances of the VSP comprises:

• generating a respective set of two or more temporal frames for each of the plurality of vocalization instances of the VSP;
• generating a corresponding set of frequency domain representations for each of the respective sets of one or more temporal frames;
• compressing the corresponding sets of frequency domain representations into respective sets of Mel frequencies having dB spectra values;
• and
• producing cepstral coefficients from the Mel frequencies.

A system provided to concurrently segment a plurality of vocalization instances of a voiced sound pattern, the system comprising:

• one or more audio sensors configured to receive an audible signal and convert the received audible signal into electronic signal data;
• a feature extraction module configured to obtain the plurality of vocalization instances of a voiced sound pattern (VSP) from the electronic signal data;
• a mapping module configured to generate a mapping between each of the plurality of vocalization instances of a VSP by performing a dynamic time warping process on corresponding spectral components of portions of the plurality of vocalization instances of the VSP, and wherein each of the plurality of vocalization instances of the VSP has one or more vocal characteristics of a particular speaker;
• and
• a multi-VSP hierarchical agglomerative clustering (HAC) module configured to concurrently segment the plurality of vocalization instances of the VSP using the mapping and in order to satisfy a performance condition using a modified HAC process.