A novel approach based on the use of acoustic simulations of impaired perception and objective measures of perceptual speech quality (PESQ) is proposed and validated to predict speech intelligibility in sensorineural hearing loss. The proposed approach consists of three steps: first, acoustic simulations (based on spectral smearing, expansive nonlinearity, and level scaling) were applied for degrading speech waveform to simulate impaired perception in normal listeners. Second, PESQ have been employed to objectively measure the perceptual quality of the degraded speech waveform obtained from the acoustic simulation. Third, PESQ scores were transformed into predicted intelligibility scores by means of a logistic function. A set of CV and VC syllables in /a/, /u/, and /i/ contexts was used as reference test material. Subjective measures of intelligibility of the degraded speech were obtained in a group of 10 subjects and were compared to speech intelligibility predicted by the proposed approach. Overall, results from this study revealed significant effects of speech characteristics (i.e., vowel context) and hearing loss severity on intelligibility of speech waveforms subjected to distortions that simulate sensorineural hearing impairment. Objective PESQ measures showed good correlation with subjective intelligibility scores, as reported by the goodness of fit of the mapping function in terms of correlation coefficient (R2=0.7) and RMSE = 0.14, revealing that the proposed approach is a valuable aid in real clinical applications.

Objective Measures of Speech Intelligibility in Sensorineural Hearing Loss

Tognola G;Moriconi S;Chiaramello E
2014-01-01

Abstract

A novel approach based on the use of acoustic simulations of impaired perception and objective measures of perceptual speech quality (PESQ) is proposed and validated to predict speech intelligibility in sensorineural hearing loss. The proposed approach consists of three steps: first, acoustic simulations (based on spectral smearing, expansive nonlinearity, and level scaling) were applied for degrading speech waveform to simulate impaired perception in normal listeners. Second, PESQ have been employed to objectively measure the perceptual quality of the degraded speech waveform obtained from the acoustic simulation. Third, PESQ scores were transformed into predicted intelligibility scores by means of a logistic function. A set of CV and VC syllables in /a/, /u/, and /i/ contexts was used as reference test material. Subjective measures of intelligibility of the degraded speech were obtained in a group of 10 subjects and were compared to speech intelligibility predicted by the proposed approach. Overall, results from this study revealed significant effects of speech characteristics (i.e., vowel context) and hearing loss severity on intelligibility of speech waveforms subjected to distortions that simulate sensorineural hearing impairment. Objective PESQ measures showed good correlation with subjective intelligibility scores, as reported by the goodness of fit of the mapping function in terms of correlation coefficient (R2=0.7) and RMSE = 0.14, revealing that the proposed approach is a valuable aid in real clinical applications.
2014
Istituto di Elettronica e di Ingegneria dell'Informazione e delle Telecomunicazioni - IEIIT
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14243/256632
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact