Speech samples of 9 subjects (8 males, 1 female) were recorded before and 0.5-2 years after a partial glossectomy and reconstruction with a pectoralis major myocutaneous flap. A reading sample, a list of meaningful and nonsense words, and a list of sustained vowels were recorded. The speech samples were evaluated by pairs of naive listeners and using acoustic analysis of the vowel production. Each pair listened to the recordings of only one patient. Inter-rater agreement was satisfactory. The general impression of the speech outcome varied from normal to moderately impaired. The perceptually estimated impairments of speech articulation in the after/before comparisons were statistically significant. Only the first formant of the vowel /i/ (rise) and the second formant of the vowel /a/ (drop) changed significantly at the group level. There was a negative correlation (r = -0.79) between the extent of tongue resection and the drop of the second formant of the vowel /a/. The perceptual variables showed a relationship (r = 0.74-0.82) with the changes in the level of the second formant of the vowel /i/. The relationship that emerged between the perceptual estimates and the objective acoustic parameters suggests that it will be possible to develop clinically relevant test batteries for articulatory quality analysis.

No abstract

No abstract

No abstract