Produção Nacional
Revisado por pares
Transfer Learning with AudioSet to Voice Pathologies Identification in Continuous Speech
2019; Elsevier BV; Volume: 164; Linguagem: Inglês
10.1016/j.procs.2019.12.233
ISSN1877-0509
AutoresVictor Guedes, Felipe Teixeira, Alessa Anjos de Oliveira, Joana Fernandes, Letícia Silva, Arnaldo Cândido, João Paulo Teixeira,
Tópico(s)Speech Recognition and Synthesis
ResumoThe 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.
Referência(s)