Abstract: An input signal may be classified by computing correlations between feature vectors of the input signal and feature vectors of reference signals, wherein the reference signals correspond to a class. The feature vectors of the input signal and/or the reference signals may be segmented to identify portions of the signals before performing the correlations. Multiple correlations of the segments may be combined to produce a segment score corresponding to a segment. The signal may then be classified using multiple segment scores, for example by comparing a combination of the segment scores to a threshold.

Abstract: Features that may be computed from a harmonic signal include a fractional chirp rate, a pitch, and amplitudes of the harmonics. A fractional chirp rate may be estimated, for example, by computing scores corresponding to different fractional chirp rates and selecting a highest score. A first pitch may be computed from a frequency representation that is computed using the estimated fractional chirp rate, for example, by using peak-to-peak distances in the frequency distribution. A second pitch may be computed using the first pitch, and a frequency representation of the signal, for example, by using correlations of portions of the frequency representation. Amplitudes of harmonics of the signal may be determined using the estimated fractional chirp rate and second pitch. Any of the estimated fractional chirp rate, second pitch, and harmonic amplitudes may be used for further processing, such as speech recognition, speaker verification, speaker identification, or signal reconstruction.

Abstract: The technology described in this document can be embodied in a computer-implemented method that includes obtaining a speech signal, and estimating a first set and a second set of segment boundaries using the speech signal. The first and second set of segment boundaries are determined using a first and second segmentation process, respectively. The second segmentation process is different from the first segmentation process. The method also includes obtaining a model corresponding to a distribution of segment boundaries, computing a first score indicative of a degree of similarity between the model and the first set of segment boundaries, and computing a second score indicating a degree of similarity between the model and the second set of segment boundaries. The method further includes selecting a set of segment boundaries using the first score and the second score, and processing the speech signal using the selected set of segment boundaries.

Abstract: An input signal may be classified by comparing a trajectory of a sequence of feature vectors of the input signal to sequences of feature vectors of reference signals, wherein the reference signals correspond to classes. For a class, a score may be computed that indicates a match between the trajectory of the input signal with trajectories of reference sequences corresponding to the class. The input signal may be classified by selecting a class corresponding to a highest score. In some implementations, the score may by computed by determining a number of nearest neighbors of the class to the input signal or by sequentially processing the input signal and updating a score for successive steps of the input sequence.