Performance of Acoustic Measures for the Discrimination Among Healthy, Rough, Breathy, and Strained Voices Using the Feedforward Neural Network.

OBJECTIVE: To identify and evaluate the best set of acoustic measures to discriminate among healthy, rough, breathy, and strained voices. METHODS: This study used the vocal samples of the sustained /ε/ vowel from 251 patients with the vocal complaints, among which 51, 80, 63, and 57 patients exhibited healthy, rough, breathy, and strained voices, respectively. Twenty-two acoustic measures were extracted, and feature selection was applied to reduce the number of combinations of acoustic measures and obtain an optimal subset of measures according to the information gain attribute ranking algorithm. To classify signals as a function of predominant voice quality, a feedforward neural network was applied using a Levenberg-Marquardt supervised learning algorithm. RESULTS: The best results were obtained from 11 combinations, with each combination presenting six acoustic measures. Kappa indices ranged from 0.7527 to 0.7743, the overall hit rates are 81.67%-83.27%, and the hit rates of healthy, rough, breathy, and strained voices are 74.51%-84.31%, 78.75%-90.00%, 85.71%-98.41%, and 68.42%-82.46%, respectively. CONCLUSIONS: We obtained the best results from 11 combinations, with each combination exhibiting six acoustic measures for discriminating among healthy, rough, breathy, and strained voices. These sets exhibited good Kappa performance and a good overall hit rate. The hit rate varied between acceptable and good for healthy voices, acceptable and excellent for rough voices, good and excellent for breathy voices, and poor and good for strained voices.

Auditory Perception of Roughness and Breathiness by Dysphonic Women.

OBJECTIVE: To investigate the auditory perception of roughness and breathiness by dysphonic women. METHODS: Twenty-two dysphonic native Brazilian Portuguese women participated in this research. All participants underwent audiological evaluation and laryngeal examination to confirm the diagnosis. During the tests, they recorded the sustained vowel /Ɛ/. A speech-language pathologist performed the auditory-perceptual judgment of voice quality for these vocal samples, categorizing the general degree of vocal deviation (mild, moderate, and severe degree) and the predominant type of deviation (roughness or breathiness). Thirty-two (32) stimuli were selected from a voice database, including twenty-four (24) dysphonic voice samples and eight (8) voice samples from vocally healthy women. The authors conducted five perception experiments, being three categorization tasks (normal vs. deviated, breathy vs. nonbreathy, rough vs. nonrough) and two tasks for discriminating the degree of deviation (roughness degree and breathiness degree). RESULTS: The experiments showed a difference between the answers for presence/absence of deviation, presence/absence of breathiness, and presence/absence of roughness in the stimuli, and a difference in the proportion of similar answers of dysphonic women (P < 0.001) regarding the identification of the deviation. Participants classified a large part of the deviated (57.9%), breathy (63.13%), and rough (65.31%) voices as normal. The degree of vocal deviation (P = 0.008) and the degree of roughness in the stimuli correlated positively with the proportion of similar answers of the participants. As for the discrimination of breathiness degrees, less deviated (normal and mild) voices were less discriminated, and more deviated (moderate and severe) voices were better discriminated. Regarding the discrimination of roughness degrees, only the voices with severe deviations showed good discrimination. CONCLUSION: Dysphonic women had a high rate of not similar answers in the identification of normal and deviated voices. They identified more than half of the deviated voices as normal. Samples with more severe deviations were proportionally more identified as deviated by the participants. The greater the vocal deviation of the participants' voices, the smallest the number of similar answers. Participants had a high rate of not similar answers in the identification of normal and breathy voices. Dysphonic women show less ability to perceive mildly and moderately breathy voices in the breathy category. Participants had a high rate of similar answers in the identification of normal and rough voices. Dysphonic women show less ability to perceive mildly and moderately breathy voices in the breathy category. Participants show less ability to perceive only mildly roughness voices with similar responses. Dysphonic women could discriminate between voices with adjacent degrees of roughness but had a low percentage of similar answers for discrimination between voices with adjacent degrees of breathiness.

Cepstral measures in the assessment of severity of voice disorders. / Medidas cepstrais na avaliação da intensidade do desvio vocal.

PURPOSE: To analyze whether there is an association between the presence, intensity and type of voice disorder and the cepstral measures in samples of individuals with voice complaints. METHODS: We used 376 vowel /Ɛ/ samples from individuals of both genders that had voice complaints. An analogue-visual scale was used for the auditory-perceptual analysis of voices regarding the overall grade of dysphonia (G) and the grades of roughness (R), breathiness (B), and strain (S), including a determination of voice quality (rough, breathy or strained). Measures related to cepstral peak prominence smoothed (CPPS) and spectral decline of vocal samples were extracted. RESULTS: There were differences in the CPPS values between the groups with or without voice disorders as well as between the different intensities and types of voice disorder. CPPS values were lower because of the presence and intensity of voice disorders. The CPPS values differentiated the following voices: rough x breathy, rough x strained, and breathy x strained. The spectral decline only differentiated breathy x strained voices. CPPS correlated positively and strongly with G and B; moderately and negatively with R, and negatively and weakly with S. The spectral decline had a moderate positive correlation with S and a weak negative correlation with B. CONCLUSION: There is association between voice disorder, G, predominant voice quality, and CPPS. In particular, G is strongly correlated with CPPS. Spectral decline is associated only with the parameters B and S.

OBJETIVO: Analisar se existe associação entre a presença, a intensidade e o tipo de desvio vocal e as medidas cepstrais em amostras de indivíduos com queixa vocal. MÉTODO: Foram utilizadas 376 amostras da vogal /ε/ de indivíduos de ambos os gêneros, com queixa vocal. Utilizou-se uma escala analógico-visual para análise perceptivo-auditiva das vozes quanto à intensidade do desvio vocal (GG), graus de rugosidade (GR), soprosidade (GS) e tensão (GT), incluindo-se a determinação da qualidade vocal predominante (rugosa, soprosa ou tensa). Foram extraídas as medidas relacionadas ao Cepstral Peak Prominence-Smoothed (CPPS) e o declínio espectral das amostras vocais. RESULTADOS: Houve diferença dos valores do CPPS entre os grupos com e sem desvio vocal, assim como entre as diferentes intensidades e tipos de desvio vocal. Os valores do CPPS foram mais reduzidos em função da presença e intensidade do desvio vocal. Os valores do CPPS diferenciaram vozes rugosas x soprosas, rugosas x tensas e soprosas x tensas. O declínio espectral apenas diferenciou vozes soprosas x tensas. O CPPS se correlacionou de modo positivo e forte com os GG e GS, de modo negativo moderado com o GR, e de forma negativa fraca com o GT. O declínio espectral apresentou correlação positiva moderada com o GT e correlação negativa fraca com o GS. CONCLUSÃO: Existe associação entre a presença de desvio vocal, o GG, a qualidade vocal predominante e o CPPS. De modo especial, o GG é fortemente correlacionado ao CPPS. O declínio espectral está associado apenas aos parâmetros de soprosidade e tensão.

Classificação espectrográfica do sinal vocal: relação com o diagnóstico laríngeo e a análise perceptivo-auditiva / Spectrographic classification of the vocal signal: relation with laryngeal diagnosis and auditory-perceptual analysis

RESUMO Objetivo Verificar se existe associação entre a presença de alteração laríngea, a análise perceptivo-auditiva da qualidade vocal e a classificação espectrográfica do sinal vocal em indivíduos com distúrbio de voz. Métodos Participaram 478 pacientes com distúrbios de voz. Foi realizada gravação da vogal /Ɛ/ sustentada e o exame médico para estabelecimento de diagnóstico laríngeo. Os espectrogramas da vogal foram utilizados para classificação dos sinais em Tipo I, II, III e IV. Resultados Vozes de indivíduos sem alteração laríngea foram classificadas, predominantemente, como Tipo I e Tipo II, enquanto sinais de indivíduos com alteração laríngea foram classificados nos Tipos III e IV. Vozes desviadas foram classificadas, predominantemente, como Tipo II, enquanto os sinais de pacientes com desvio vocal foram categorizados, predominantemente, como Tipos II e III. Apenas os sinais de indivíduos com desvio vocal foram classificados como Tipo IV. Sinais Tipo III e IV apresentaram valores mais elevados no grau geral do desvio e nos graus de rugosidade e soprosidade, em relação aos sinais Tipo I e Tipo II. Os sinais Tipo IV apresentaram maior grau geral e graus de rugosidade e soprosidade, em comparação aos sinais Tipo III. Apenas os sinais Tipo IV apresentaram valores mais elevados no grau de tensão, em relação aos sinais Tipo I, II e III. Conclusão Há associação entre a presença de alteração laríngea, a análise perceptivo-auditiva e a classificação espectrográfica do sinal vocal em indivíduos com distúrbio de voz.

ABSTRACT Purpose To verify whether there is an association between the presence of laryngeal alteration, auditory-perceptual analysis of vocal quality, and the spectrographic classification of the vocal signal in individuals with voice disorders. Methods 478 patients with voice disorders participated in the study. A recording of the sustained vowel /Ɛ/ and a medical examination were performed to establish a laryngeal diagnosis. The vowel spectrograms were used to classify the signals into type I, II, III and IV. Results Voices of individuals without laryngeal disorders were predominantly classified as type I and type II, while signals of individuals with laryngeal disorders were classified as types III and IV. Deviated voices were predominantly classified as type II, while the signals of patients with vocal deviation were predominantly categorized as types II and III. Only the signals of individuals with vocal deviation were classified as type IV. Type III and IV signals showed higher values for the general degree of deviation and for degrees of roughness and breathiness in relation to type I and type II signals. Type IV signals showed a higher general degree and degrees of roughness and breathiness compared to type III signals. Only type IV signals showed higher values in the degree of tension compared to types I, II and III signals. Conclusion There is an association between the presence of laryngeal alteration, auditory-perceptual analysis, and the spectrographic classification of the vocal signal in individuals with voice disorders.

Medidas cepstrais na avaliação da intensidade do desvio vocal / Cepstral measures in the assessment of severity of voice disorders

RESUMO Objetivo Analisar se existe associação entre a presença, a intensidade e o tipo de desvio vocal e as medidas cepstrais em amostras de indivíduos com queixa vocal. Método Foram utilizadas 376 amostras da vogal /ε/ de indivíduos de ambos os gêneros, com queixa vocal. Utilizou-se uma escala analógico-visual para análise perceptivo-auditiva das vozes quanto à intensidade do desvio vocal (GG), graus de rugosidade (GR), soprosidade (GS) e tensão (GT), incluindo-se a determinação da qualidade vocal predominante (rugosa, soprosa ou tensa). Foram extraídas as medidas relacionadas ao Cepstral Peak Prominence-Smoothed (CPPS) e o declínio espectral das amostras vocais. Resultados Houve diferença dos valores do CPPS entre os grupos com e sem desvio vocal, assim como entre as diferentes intensidades e tipos de desvio vocal. Os valores do CPPS foram mais reduzidos em função da presença e intensidade do desvio vocal. Os valores do CPPS diferenciaram vozes rugosas x soprosas, rugosas x tensas e soprosas x tensas. O declínio espectral apenas diferenciou vozes soprosas x tensas. O CPPS se correlacionou de modo positivo e forte com os GG e GS, de modo negativo moderado com o GR, e de forma negativa fraca com o GT. O declínio espectral apresentou correlação positiva moderada com o GT e correlação negativa fraca com o GS. Conclusão Existe associação entre a presença de desvio vocal, o GG, a qualidade vocal predominante e o CPPS. De modo especial, o GG é fortemente correlacionado ao CPPS. O declínio espectral está associado apenas aos parâmetros de soprosidade e tensão.

ABSTRACT Purpose To analyze whether there is an association between the presence, intensity and type of voice disorder and the cepstral measures in samples of individuals with voice complaints. Methods We used 376 vowel /Ɛ/ samples from individuals of both genders that had voice complaints. An analogue-visual scale was used for the auditory-perceptual analysis of voices regarding the overall grade of dysphonia (G) and the grades of roughness (R), breathiness (B), and strain (S), including a determination of voice quality (rough, breathy or strained). Measures related to cepstral peak prominence smoothed (CPPS) and spectral decline of vocal samples were extracted. Results There were differences in the CPPS values between the groups with or without voice disorders as well as between the different intensities and types of voice disorder. CPPS values were lower because of the presence and intensity of voice disorders. The CPPS values differentiated the following voices: rough x breathy, rough x strained, and breathy x strained. The spectral decline only differentiated breathy x strained voices. CPPS correlated positively and strongly with G and B; moderately and negatively with R, and negatively and weakly with S. The spectral decline had a moderate positive correlation with S and a weak negative correlation with B. Conclusion There is association between voice disorder, G, predominant voice quality, and CPPS. In particular, G is strongly correlated with CPPS. Spectral decline is associated only with the parameters B and S.