We propose a mask-based enhancer for very low quality speech that is able to preserve important cues in a noise-robust manner by identifying the time-frequency regions that contain significant speech energy. We use a classifier to estimate a time-frequency mask from an input feature set that provides information about the energy distribution of both voiced and unvoiced speech. We evaluate the enhancer on a range of noisy speech signals and demonstrate that it yields consistent improvements in an objective intelligibility measure.

We present PEFAC, a fundamental frequency estimation algorithm for speech that is able to identify voiced frames and estimate pitch reliably even at negative signal-to-noise ratios. The algorithm combines a normalization stage, to remove channel dependency and to attenuate strong noise components, with a harmonic summing filter applied in the log-frequency power spectral domain, the impulse response of which is chosen to sum the energy of the fundamental frequency harmonics while attenuating smoothly-varying noise components. Temporal continuity constraints are applied to the selected pitch candidates and a voiced speech probability is computed from the likelihood ratio of two classifiers, one for voiced speech and one for unvoiced speech/silence. We compare the performance of our algorithm with that of other widely used algorithms and demonstrate that it performs well in both high and low levels of additive noise.

In this paper, we present a novel approach to the classification of speaker likability, that is, a measure of how pleasant a given speaker is to listen to. Instead of blindly extracting a large number of features, we identify a small set of features which represent perceptual speech characteristics. This set of features is sent to a linear support vector machine to perform speaker likability classification. We train and evaluate the performance of our algorithm on the Interspeech 2012 speaker trait challenge database and we show that our likability classifier achieves an absolute improvement of 3.2% over the baseline classifier developed for the challenge while considerably reducing the number of features needed.

We present a new method for speech active level estimation which combines a novel algorithm based on voiced speech energy extraction with the standardized ITU-T Recommendation P.56. At poor signal-to-noise ratios, the algorithm estimates the active level by identifying intervals of voiced speech and summing the energy of the pitch harmonics in the time-frequency domain while rejecting that of the noise. We compare the performance of our method with that of ITU-T P.56 on the TIMIT database and demonstrate that it performs exceptionally well in both high and low levels of additive noise.

We present an algorithm for identifying the location of sibilant phones in noisy speech. Our algorithm does not attempt to identify sibilant onsets and offsets directly but instead detects a sustained increase in power over the en-tire duration of a sibilant phone. The normalized esti-mate of the sibilant power in each of 14 frequency bands forms the input to two Gaussian mixture models that are trained on sibilant and non-sibilant frames respectively. The likelihood ratio of the two models is then used to classify each frame. We evaluate the performance of our algorithm on the TIMIT database and demonstrate that the classification accuracy is over 80 % at 0 dB signal to noise ratio for additive white noise. Index Terms: sibilant speech, spectrographic mask esti-mation, speech classification, speech segregation 1.