Browsing by Author "Oliveira, Alessa Anjos de"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
- Clustering of voice pathologies based on sustained voice parametersPublication . Oliveira, Alessa Anjos de; Dajer, Maria; Fernandes, Paula Odete; Teixeira, João PauloSignal processing techniques can be used to extract information that contribute to the detection of laryngeal disorders. The goal of this paper is to perform a statistical analysis through the boxplot tool from 832 voice signals of individuals with different laryngeal pathologies from the Saarbrücken Voice Database in order to create relevant groups, making feasible an automatic identification of these dysfunctions. Jitter, Shimmer, HNR, NHR and Autocorrelation features were compared between several groups of voice pathologies/conditions, resulting in three identified clusters.
- Clustering pathologic voice with kohonen SOM and hierarchical clusteringPublication . Teixeira, João Paulo; Dajer, Maria; Oliveira, Alessa Anjos deThe main purpose of clustering voice pathologies is the attempt to form large groups of subjects with similar pathologies to be used with Deep-Learning. This paper focuses on applying Kohonen's Self-Organizing Maps and Hierarchical Clustering to investigate how these methods behave in the clustering procedure of voice samples by means of the parameters absolute jitter, relative jitter, absolute shimmer, relative shimmer, HNR, NHR and Autocorrelation. For this, a comparison is made between the speech samples of the Control group of subjects, the Hyper-functional Dysphonia and Vocal Folds Paralysis pathologies groups of subjects. As a result, the dataset was divided in two clusters, with no distinction between the pre-defined groups of pathologies. The result is aligned with previous result using statistical analysis.
- Exploração de técnicas de clustering aplicadas a patalogias da vozPublication . Oliveira, Alessa Anjos de; Teixeira, João Paulo; Dajer, Maria EugêniaEste trabalho de dissertação de mestrado tem o intuito de aplicar técnicas de clustering em um conjunto de dados a fim de verificar a possibilidade de agrupamento entre grupos que contém informações a respeito da voz. Neste estudo são empregados os parâmetros acústicos jitter absoluto, jitter relativo, shimmer relativo, autocorreleção, Harmonic to Noise Ratio e Noise to Harmonic Ratio. Estes atributos são extraídos da base de dados Saarbrücken Voice Database, a partir de ficheiros de áudio que dispõem das vogais /a/, /i/e/u/ sustentadas nos tons alto, baixo e normal. Para o agrupamento dos conjuntos através desses parâmetros são utilizados o grupo de Controlo, o Carcinoma, o Cisto, a Disfonia Espasmódica, a Disfunção Funcional, a Disfunção Psicogênica, o Edema de Reinke, o Granuloma, o Granuloma por Intubação. a Laringite Crônica, a Paralisia das Cordas Vocais, Pólipos das Cordas Vocais, o Tumorna Hipofaringe e o Tumor na Laringe. Esta invetigação ocorre ao empregar 2 caminhos diferentes para averuiguar a possibilidade de agrupar patalogias: a análise estatística descritiva e algoritmos de agrupamento. Para a análise estatística descritiva, utiliza-se o Diagrama de Caixas. Para os métodos de agrupamento, são empregados os Mapas Auto- Organizáveis de Kohonen e o clustering hierárquico aglomerativo. Os resultados dessas categorias apontam a divisão dos dados em 2 conjuntos, em que um grupo contém grande parte das amostras, enquanto o outro contém apenas algumas amostras. No entanto, essas técnicas se divergem ao comparar os elementos pertencentes aos conjuntos, pois, na análise estatística descritiva as amostras se encontram maia bem separadasd do que nos métodos de clustering.
- Transfer learning with audioSet to voice pathologies identification in continuous speechPublication . Guedes, Victor; Teixeira, Felipe; Oliveira, Alessa Anjos de; Fernandes, Joana Filipa Teixeira; Silva, Letícia; Candido Junior, Arnaldo; Teixeira, João PauloThe classification of pathological diseases with the implementation of concepts of Deep Learning has been increasing considerably in recent times. Among the works developed there are good results for the classification in sustained speech with vowels, but few related works for the classification in continuous speech. This work uses the German Saarbrücken Voice Database with the phrase “Guten Morgen, wie geht es Ihnen?” to classify four classes: dysphonia, laryngitis, paralysis of vocal cords and healthy voices. Transfer learning concepts were used with the AudioSet database. Two models were developed based on Long-Short-Term-Memory and Convolutional Network for classification of extracted embeddings and comparison of the best results, using cross-validation. The final results allowed to obtaining 40% of f1-score for the four classes, 66% f1-score for Dysphonia x Healthy, 67% for Laryngitis x healthy and 80% for Paralysis x Healthy.