Linear Prediction of Speech

Markel, John E.; Gray, Alfred

doi:10.1007/978-3-642-66286-7

Cited by 1,836 publications

(535 citation statements)

References 0 publications

Supporting

Mentioning

490

Contrasting

Unclassified

Order By: Relevance

“…We studied formants (F1, F2, F3, F4) using linear predictive coding (LPC; Markel and Gray 1976). LPC is based on the assumption that the vocal signal is produced by a buzz generated at the glottis.…”

Section: Sound Recording and Analysismentioning

confidence: 99%

Vocal Tract Morphology Determines Species-Specific Features in Vocal Signals of Lemurs (Eulemur)

2012

View full text Add to dashboard Cite

The source-filter theory describes vocal production as a two-stage process involving the generation of a sound source, with its own spectral structure, which is then filtered by the resonant properties of the vocal tract. This theory has been successfully applied to the study of animal vocal signals since the 1990s. As an extension, models reproducing vocal tract resonance can be used to reproduce formant patterns and to understand the role of vocal tract filtering in nonhuman vocalizations. We studied three congeneric lemur species -Eulemur fulvus, E. macaco, E. rubriventer-using morphological measurements to build computational models of the vocal tract to estimate formants, and acoustic analysis to measure formants from natural calls. We focused on call types emitted through the nose, without apparent articulation. On the basis of anatomical measurements, we modeled the vocal tract of each species as a series of concatenated tubes, with a cross-sectional area that changed along the tract to approximate the morphology of the larynx, the nasopharyngeal cavity, the nasal chambers, and the nostrils. For each species, we calculated the resonance frequencies in 2500 randomly generated vocal tracts, in which we simulated intraspecific length and size variation. Formant location and spacing showed significant species-specific differences determined by the length of the vocal tract. We then measured formants of a set of nasal vocalizations ("grunts") recorded from captive lemurs of the same species. We found species-specific differences in the natural calls. This is the first evidence that morphology of the vocal tract is relevant in generating filter-related acoustic cues that potentially provide receivers with information about the species of the emitter.

show abstract

Section: Sound Recording and Analysismentioning

confidence: 99%

Vocal Tract Morphology Determines Species-Specific Features in Vocal Signals of Lemurs (Eulemur)

2012

View full text Add to dashboard Cite

show abstract

“…We extract features related to statistics of formant contours (Table 1), pitch contours (Table 2), and energy contours ( Table 3). The method to estimate formants relies on the linear prediction analysis [35], whereas pitch is computed based on an autocorrelation method. Statistics related to the distribution of energy into several spectral bands (Table 4) as well as statistics of the TEO-autocorrelation (Table 5) are also computed.…”

Section: Feature Extractionmentioning

confidence: 99%

Speaker-independent emotion recognition exploiting a psychologically-inspired binary cascade classification schema

Kotti

Paternò

2012

Int J Speech Technol

View full text Add to dashboard Cite

In this paper, a psychologically-inspired binary cascade classification schema is proposed for speech emotion recognition. Performance is enhanced because commonly confused pairs of emotions are distinguishable from one another. Extracted features are related to statistics of pitch, formants, and energy contours, as well as spectrum, cepstrum, perceptual and temporal features, autocorrelation, MPEG-7 descriptors, Fujisaki's model parameters, voice quality, jitter, and shimmer. Selected features are fed as input to K nearest neighborhood classifier and to support vector machines. Two kernels are tested for the latter: linear and Gaussian radial basis function. The recently proposed speaker-independent experimental protocol is tested on the Berlin emotional speech database for each gender separately. The best emotion recognition accuracy, achieved by support vector machines with linear kernel, equals 87.7%, outperforming state-of-the-art approaches. Statistical analysis is first carried out with respect to the classifiers' error rates and then to evaluate the information expressed by the classifiers' confusion matrices.

show abstract

“…Instead, for each phoneme all patients were treated as one group and independent factors were the two repeated measurement interviews (AI and DI) and repetition of the utterances (three times). Dependent variables were the formant frequencies which were extracted by means of linear prediction routines (Markel and Gray, 1976).…”

Section: Designmentioning

confidence: 99%

Vocal indicators of psychiatric treatment effects in depressives and schizophrenics

Tolkmitt

Helfrich

Standke

et al. 1982

Journal of Communication Disorders

View full text Add to dashboard Cite

Voice and speech changes as a result of clinical treatment for 17 depressive and 15 schizophrenic patients were investigated.Speech samples taken at interviews before and after treatment were analyzed with regard tofc, spectral energy distribution, and formant frequencies of vowels occurring in identical phonetical context. Both groups of patients showed a decrease in f. after therapy, which was interpreted as a reduction of general arousal. Differential results with regard to spectral energy distribution suggested that the voice of depressives became more relaxed after therapy, but the opposite seemed to be true for schizophrenics.

show abstract

Linear Prediction of Speech

Cited by 1,836 publications

References 0 publications

Vocal Tract Morphology Determines Species-Specific Features in Vocal Signals of Lemurs (Eulemur)

Vocal Tract Morphology Determines Species-Specific Features in Vocal Signals of Lemurs (Eulemur)

Speaker-independent emotion recognition exploiting a psychologically-inspired binary cascade classification schema

Vocal indicators of psychiatric treatment effects in depressives and schizophrenics

Contact Info

Product

Resources

About