COMPUTER SIMULATION OF MUSICAL SINGER'S VOICE BASED ON MRI AND ACOUSTIC MEASUREMENTS

An inverse method was used to estimate the vocal tract geometry as a 1 D model on the basis of acoustical characteristics of a professional musical singer before and after vocal exercising. The basic geometrical data for the model were obtained from magnetic resonance images (MRI) registered during sustained phonation of vowels (a:), (i:), (u:) produced in naive and professional ways (before and after exercising respectively). The model was used for numerical simulations of the voice signals. The results of simulation were compared to the acoustic recordings. According to the results, a singer's formant cluster was accomplished after exercising. It seemed to be due to lowering of the larynx and lengthening and narrowing of the epilarynx. The area ratio between the low pharynx and the epilarynx increased for (a:) but decreased for (i:) and (u:) after exercising being between 3.33 and 4.39. There was a qualitative agreement between 3D measurements of MRI and the results of modeling. The results suggest that for a singer's formant cluster a relatively low pharynx over epilarynx ratio may be sufficient, at least if the larynx lowers. Sundberg (1974) formulated an acoustic interpretation of singers formant as clustering of formants F3 - F5 in the vicinity of 3 kHz. According to the calculations by Fant (1960) such a situation can occur if the ratio of the cross-sectional area of the lower pharynx over the outlet of the epilaryngeal tube is six or higher. The physical modeling results by Sundberg supported the findings of Fant. Different results were obtained by Detweiler (1994) who investigated by MRI scanning the vowel phonation of tenor and baritone singers phonating vowel (a:) in modal register. According to her results, the subjects laryngopharyngeal/laryngeal outlet cross-sectional area ratios varied from 2.9:1 to 3.7:1, which wasnt consistent with the hypotheses by Sundberg and Fant. Later, Sundberg (2003) criticized the Detweiler's way of measuring the areas, which was accomplished from the sagittal and transversal distances measured in 2D slices. The epilarynx tube, located just above the vocal folds, was also theoretically considered as a dominant source of clustering of formant frequencies especially when the cross-sectional area of this tube is approximately six times less than the area of the lower pharynx by Titze and Story (1997), and Story (2003). However, the physiological adjustment used in singers or in actors has not been known in sufficient details yet.