Spectral pattern recognition of improved voice quality

Rihkanen, Heikki; Leinonen, Lea; Hiltunen, Tapio; Kangas, Jari

doi:10.1016/s0892-1997(05)80280-2

Cited by 10 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies do not seem to support these findings, since most authors consider the principal predictions of breathy voice to be presence of aspiration noise [2,25,33,66], together with energy losses and gains in various spectral zones [9,67].…”

Section: Acoustic Correlates Of Vocal Qualitymentioning

confidence: 86%

Acoustic and Perceptual Indicators of Normal and Pathological Voice

Muñoz

Mendoza

Fresneda

et al. 2003

Folia Phoniatr Logop

View full text Add to dashboard Cite

The aim of this study was to obtain acoustic correlates to vocal quality of a group of men and women with and without voice disorders, based on evaluations of a group of judges experienced in the field of vocal rehabilitation. In male subjects, perceptual evaluation of normal, hoarse and rough voice qualities was related to the following acoustic features: frequency perturbation measures (JITA, RAP, and SPPQ), amplitude perturbation (SAPQ and VAM), soft phonation index (SPI) and fundamental frequency tremor intensity (FTRI). While these measures presented normal values for normal voice, hoarseness showed some deviations in perturbation frequency variables and very high SPI values, while rough voice showed deviations in all the measures. Qualities of female voices were perceived as normal, breathy and hoarse, but the acoustic correlates of these qualities were less conclusive.

show abstract

Section: Acoustic Correlates Of Vocal Qualitymentioning

confidence: 86%

Acoustic and Perceptual Indicators of Normal and Pathological Voice

Muñoz

Mendoza

Fresneda

et al. 2003

Folia Phoniatr Logop

View full text Add to dashboard Cite

show abstract

“…Mean, maximum and minimum F0 were computed for the speech samples and part of the monkey samples as previously described (52). Due to the noise, 'commanding', 'angry', and 'frightened' samples could not be automatically analysed.…”

Section: Acoustic Analysesmentioning

confidence: 99%

“…In monkeys, the audible spectrum may extend beyond 30 kHz (56). For the visualisation of differences in the overall distribution of spectral energy, selforganised maps were computed (30,35,52) using spectral features from the initial 150 ms of the signal (Fig. 3).…”

Section: Acoustic Analysesmentioning

confidence: 99%

Shared means and meanings in vocal expression of man and macaque

Leinonen

Laakso²,

Carlson

et al. 2003

Logopedics Phoniatrics Vocology

Self Cite

View full text Add to dashboard Cite

Vocalisations of six Macaca arctoides that were categorised according to their social context and judgements of naive human listeners as expressions of plea/submission, anger, fear, dominance, contentment and emotional neutrality, were compared with vowel samples extracted from simulations of emotional-motivational connotations by the Finnish name Saara and English name Sarah. The words were spoken by seven Finnish and 13 English women. Humans and monkeys resembled each other in the following respects. 'Neutral', 'pleading' and 'commanding' had a similar F0 level. Loud vocalisations of intense 'anger' and 'fear' had both high F0, and the highest values were encountered for 'fear'. Compared with 'neutral', the audiosignal waveform of 'plea/submission' was more sinusoidal, seen in the spectrum as an attenuation of formants and an emphasis of the fundamental, whereas the signal waveform of 'commanding' was more complex corresponding to an increase in noise and a wider distribution of spectral energy. 'Frightened' samples included rather harmonic segments with emphasis of the fundamental, and 'angry' samples included more noise at the low end of the spectrum and often segments with low-frequency ( < 100 Hz) edged modulation. Sounds resembling soft and noisy 'content' grunts of the monkey do not appear in Finnish or English speech but 'content' utterances were, however, associated with low speech pressure, attenuation of harmonics and increase in noise.

show abstract

“…SOMs have been used to characterize several aspects of normal and disordered voice. In a series of studies carried out by Leinonen and colleagues (Leinonen, Hiltunen, Kangas, Juvas, & Rihkanen, 1993;Leinonen, Kangas, Torkkola, & Juvas, 1992;Rihkanen, Leinonen, Hiltunen, & Kangas, 1994), a SOM was trained to detect characteristics of dysphonic voice by recognition of spectral composition. To determine how this map would characterize disordered voice, samples were taken from individuals classified with varying degrees of dysphonia using a method derived from the GRBAS perceptual rating scale (Hirano, 1981).…”

mentioning

confidence: 99%

Self-Organizing Map for the Classification of Normal and Disordered Female Voices

Callan

Kent

Roy

et al. 1999

J Speech Lang Hear Res

View full text Add to dashboard Cite

The goal of this research was to train a self-organizing map (SOM) on various acoustic measures (amplitude perturbation quotient, degree of voice breaks, rahmonic amplitude, soft phonation index, standard deviation of the fundamental frequency, and peak amplitude variation) of the sustained vowel /a/ to enhance visualization of the multidimensional nonlinear regularities inherent in the input data space. The SOM was trained using 30 spasmodic dysphonia exemplars, 30 pretreatment functional dysphonia exemplars, 30 post-treatment functional dysphonia exemplars, and 30 normal voice exemplars. After training, the classification performance of the SOM was evaluated. The results indicated that the SOM had better classification performance than that of a stepwise discriminant analysis over the original data. Analysis of the weight values across the SOM, by means of stepwise discriminant analysis, revealed the relative importance of the acoustic measures in classification of the various groups. The SOM provided both an easy way to visualize multidimensional data, and enhanced statistical predictability at distinguishing between the various groups (over that conducted on the original data set). We regard the results of this study as a promising initial step into the use of SOMs with multiple acoustic measures to assess phonatory function.

show abstract

Spectral pattern recognition of improved voice quality

Cited by 10 publications

References 18 publications

Acoustic and Perceptual Indicators of Normal and Pathological Voice

Acoustic and Perceptual Indicators of Normal and Pathological Voice

Shared means and meanings in vocal expression of man and macaque

Self-Organizing Map for the Classification of Normal and Disordered Female Voices

Contact Info

Product

Resources

About