Evaluation of the progress of therapy in patients with cleft lip, jaw and palate, using voice analysis--a pilot study

The defective morphology of the hard and soft palate in patients with cleft lip, jaw and/or palate in conjunction with speech, voice and hearing disorders can considerably restrict the ability to communicate. Changes in vocal timbre and the centripetal displacement of articulation are characteristic of cleft palate speech. There has not been a uniform diagnostic method in cleft centers to date which makes possible the analysis, documentation and comparison of changes in timbre. In this study we assessed a computer-aided evaluation process to determine objective vocal timbre parameters while treating children with cleft lip, jaw and palate who had undergone surgery according to the principle of delayed palate repair.

The data and findings of 24 patients with various cleft forms were evaluated. The group we investigated consisted of seven patients with bilateral clefts, fourteen with unilateral cleft, and three with isolated cleft palate. Our subject cohort was limited to those born between 1985 and 1986. At the baseline investigation the children were aged 3 to 4 years, at the second investigation 4 to 5 years, and at the third investigation 5 to 7 years. These children underwent palate repair together with velopharyngoplasty between their fourth and fifth years. Our data were based on the tape recordings taken at these three investigation time points (before velopharyngoplasty, after velopharyngoplasty, and in the year they started school). The German words "Ball", "Kaffeekanne", "Schuhe" and "Schokoladenpudding" and the sentence "Meine Puppe heisst Sabine." were analyzed using "Multi Speech" software. The individual vowels were manually extracted from these recordings and the fundamental frequency (F0) and frequency of the first formant (F1) determined

We were able to evaluate the spectra of the vowels /a/, /i/ and /u/ of 24 children in all. There were statistically-significant differences in fundamental frequency when considering the intra-individual progress of patients with bilateral cleft lip, jaw and palate. The fundamental frequency at the second and third investigation time points was significantly lower (p=0.003; p=0.000) than that at the baseline investigation. We observed no significant differences regarding that parameter at the various time points when evaluating the children with unilateral cleft lip, jaw and palate. Careful appraisal of the individual vowels showed that the speech results correlated with the severity of the type of cleft.

Since the fundamental frequency and analysis of the first formant have proven to be suitable parameters for the analytical description of the vocal timbre of cleft patients, our results provide a solid basis for further studies. We provided evidence that this investigatory method is also effective when considering time and equipment requirements.