Application of surface electromyography (sEMG) in investigating voice production

Abstract

Voice production plays an important role in human communications, which is considered to be assisted by complex interactions among the intrinsic and perilaryngeal muscles but without a consensus. The technique of surface electromyography (sEMG) has been used in many studies to measure the perilaryngeal muscle activities related to voice production. The time-domain features (amplitude analysis) of sEMG have been used to measure dysphonia-related muscle activities. However, an integrative review in the present thesis revealed that the level of evidence supporting the use of sEMG in identifying dysphonia is not high. The frequency-domain features of the sEMG signals might serve as objective measurements of vocal fatigue, but not many studies addressed this issue. To further explore the application of sEMG in investigating voice production, three studies were conducted concerning the neuromuscular physiology of the perilaryngeal muscles in pitch and loudness control, and the use of sEMG in identifying dysphonia and vocal fatigue. First, a pilot study was conducted to determine the contributions of the perilaryngeal muscles in pitch and loudness change during phonation among vocally healthy populations. The findings showed that the suprahyoid muscle activities were significantly reduced when producing lower pitches and intensities compared to the natural baselines. The production of sustained /i/ required significantly more suprahyoid muscle activities than those of /a/ and /u/. The sternocleidomastoid (SCM) muscles did not show much sEMG activity in any of the pitch and loudness levels. The findings also showed a tendency for bilateral asymmetry in the use of suprahyoid and SCM muscles. Second, a study (Main Study 1) was conducted to investigate if sEMG could be used in identifying dysphonia using the time-domain feature (amplitude analysis) of the sEMG signals. The findings showed that dysphonic individuals, classified by either the individuals’ self-assessment or clinician-based auditory-perceptual voice quality judgment, tended to show more imbalanced suprahyoid muscle activities in voice production compared with the non-dysphonic groups. The combination of the sEMG measures from both the left and right suprahyoid muscles could be used as a predictor of dysphonia with a fair level of confidence (69.66%). Third, a second study (Main Study 2) was conducted to investigate whether sEMG could be used in identifying vocal fatigue using the frequency-domain feature of the sEMG signals. The findings showed that the median frequency of the sEMG signals (overall reduction of about 4.2 Hz) could be used to identify and quantify vocal fatigue contributed by the perilaryngeal muscles following a vocal loading task. In addition, such fatigue in perilaryngeal muscles, as far as sEMG activities are concerned, could last for at least 20 min. The findings of the present thesis supported that sEMG is valuable in (a) determining the role of perilaryngeal muscles in pitch and loudness control during phonations; (b) identifying individuals with dysphonia; and (c) identifying vocal fatigue induced by a vocal loading task.