Pathological Likelihood Index as a Measurement of the Degree of Voice Normality and Perceived Hoarseness

Godino-Llorente, Juan Ignacio; Gómez-Vilda, Pedro; Cruz–Roldán, Fernando; Blanco–Velasco, Manuel; Fraile, Rubén

doi:10.1016/j.jvoice.2009.04.003

Cited by 28 publications

(18 citation statements)

References 32 publications

(49 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To propose results comparable with the previously published works, we analyze voice recordings of sustained phonation of the vowel /a/ as well. However, unlike the previous works, we analyze a larger dataset composed of 4 different databases, namely: MEEI [18,39], SVD [62,44,44,2], AVPD [41,44] (these databases are commonly used by the community), and Príncipe de Asturias Database (PDA) [20,8,39]. Furthermore, to propose models capable of robust voice pathology detection, we do not restrict the dataset to only a subset of common vocal pathologies.…”

Section: Introductionmentioning

confidence: 99%

Towards robust voice pathology detection

Harár

Galáž

Alonso-Hernández

et al. 2018

Neural Comput & Applic

View full text Add to dashboard Cite

Automatic objective non-invasive detection of pathological voice based on computerized analysis of acoustic signals can play an important role in early diagnosis, progression tracking and even effective treatment of pathological voices. In search towards such a robust voice pathology detection system we investigated 3 distinct classifiers within supervised learning and anomaly detection paradigms. We conducted a set of experiments using a variety of input data such as raw waveforms, spectrograms, mel-frequency cepstral coefficients (MFCC) and conventional acoustic (dysphonic) features (AF). In comparison with previously published works, this article is the first to utilize combination of 4 different databases comprising normophonic and pathological recordings of sustained phonation of the vowel /a/ unrestricted to a subset of vocal pathologies. Furthermore, to our best knowledge, this article is the first to explore gradient boosted trees and deep learning for this application. The following best classification performances measured by F1 score on dedicated test set were achieved: XGBoost (0.733) using AF and MFCC, DenseNet (0.621) using MFCC, and Isolation Forest (0.610) using AF. Even though these results are of exploratory character, conducted experiments do show promising potential of gradient boosting and deep learning methods to robustly detect voice pathologies.

show abstract

Section: Introductionmentioning

confidence: 99%

Towards robust voice pathology detection

Harár

Galáž

Alonso-Hernández

et al. 2018

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…We used dataset generated at The Príncipe de Asturias (PdA) Hospital in Alcalá deHenares of Madrid [17] [29]. PdA consists of recordings of the sustained vowel /a/, with the first and last part of the utterance removed to avoid onset and offset effects [16] .…”

Section: Dataset Descriptionmentioning

confidence: 99%

Feature enhancement for classifier optimization and dimensionality reduction

Shilaskar¹,

Ghatol²

2014

2014 Annual IEEE India Conference (INDICON)

View full text Add to dashboard Cite

Voice is important for professionals like speakers, teachers, actors, singers and it is the important tool for communication. Laryngeal pathologies induce perturbations in the speech signal. Speech signal is discriminated as pathological or healthy based on roughness -breathiness -hoarseness (RBH) in the quality of signal. In recent years pattern recognition along with various signal processing techniques has emerged as an effective non invasive tool for diagnosis of pathological condition. Signal processing techniques tend to generate large number of features representing the signal. Automatic feature reduction techniques are vital in identifying the relevant features and eliminating the redundant ones. We extract features from speech signal using the acoustic analysis. Features are enhanced by alleviating gender bias. Periodic variations in the signal are captured using statistical techniques. We investigate intelligent system to generate reduced feature subset with improvement in diagnostic performance.

show abstract

“…(16,17). Autorzy korzystali z bazy komercyjnych próbek ludzkich głosów patologicznych i prawidło-Tabela 2.…”

Section: Wynikiunclassified

Nonlinear Acoustic Analysis in the Evaluation of Occupational Voice Disorders

Niebudek-Bogusz¹,

Grygiel²,

Strumiłło³

et al. 2013

Med Pr

View full text Add to dashboard Cite

StreszczenieWprowadzenie: W ostatnich latach podkreśla się, że tworzenie głosu podlega również procesom nieliniowym, powodującym aperiodyczne drgania fałdów głosowych. Drgania te nie zawsze można scharakteryzować za pomocą konwencjonalnych parametrów akustycznych, takich jak wskaźniki perturbacji częstotliwości i amplitudy. Z tego powodu wzrasta zainteresowanie nieliniowymi metodami analizy akustycznej. Celem pracy była ocena możliwości zastosowania nieliniowej analizy cepstralnej z wyznaczaniem współczynników cepstralnych (mel frequency cepstral coefficients -MFCC) w diagnozowaniu zawodowych zaburzeń głosu. Materiał i metody: Badaniami objęto 275 próbek głosów patologicznych (głoski "a" w przedłużonej fonacji oraz 4 standaryzowanych zdań) rejestrowanych u nauczycielek z zawodowo uwarunkowanymi łagodnymi zmianami przerostowymi fałdów głosowych (tj. guzkami głosowymi, polipami) oraz 200 próbek głosów prawidłowych kobiet z grupy kontrolnej. Średnia wieku grupy badanej wynosiła 45 lat, a grupy kontrolnej -43 lata. W badaniach wykonano analizę próbek głosowych za pomocą nieliniowej analizy akustycznej, w której zastosowano parametry MFCC. Wyniki: Po klasyfikacji MFCC przy pomocy odwzorowania Sammona oraz tzw. Maszyny Wektorów Nośnych uzyskano dużą trafność badanej metody. W testach wykonanych dla 475 zarejestrowanych próbek głosu zaburzenia głosu zostały wykryte z 91-procentową czułością i 83-procentową specyficznością dla głoski "a", a dla badanych zdań z czułością i specyficznością w granicach 87-100%. Wnioski: Nieliniowa analiza akustyczna z zastosowaniem współczynników MFCC może być przydatnym, obiektywnym narzędziem potwierdzającym zmiany patologiczne głośni o podłożu zawodowym. Prowadzone są dalsze badania w tym kierunku. Med. Pr. 2013;64(1):29-35 Słowa kluczowe: nieliniowa analiza akustyczna, zawodowe zaburzenia głosu, łagodne zmiany przerostowe fałdów głosowych, ŁZFG Abstract Background: Over recent years numerous papers have stressed that production of voice is subjected to the nonlinear processes, which cause aperiodic vibrations of vocal folds. These vibrations cannot always be characterized by means of conventional acoustic parameters, such as measurements of frequency and amplitude perturbations. Thus, special attention has recently been paid to nonlinear acoustic methods. The aim of this study was to assess the applicability of nonlinear cepstral analysis, including the evaluation of mel cepstral coefficients (MFCC), in diagnosing occupational voice disorders. Material and methods: The study involved 275 voice samples of pathologic voice (sustained vowel "a" and four standardized sentences) registered in female teachers with the occupation-related benign vocal fold masses (BVFM), such as vocal nodules, polyps, and 200 voice samples of normal voices from the control group of females. The mean age of patients and controls was similar (45 vs. 43 years). Voice samples from both groups were analyzed, including MFCC evaluation. Results: MFCC classification using the Sammon Mapping and Support Vector Machines...

show abstract

Pathological Likelihood Index as a Measurement of the Degree of Voice Normality and Perceived Hoarseness

Cited by 28 publications

References 32 publications

Towards robust voice pathology detection

Towards robust voice pathology detection

Feature enhancement for classifier optimization and dimensionality reduction

Nonlinear Acoustic Analysis in the Evaluation of Occupational Voice Disorders

Contact Info

Product

Resources

About