Full text
| |
| Author(s): |
Diana Conceição da Rocha
Total Authors: 1
|
| Document type: | Master's Dissertation |
| Press: | Bauru. |
| Institution: | Universidade de São Paulo (USP). Faculdade de Odontologia de Bauru (FOB/SDB) |
| Defense date: | 2016-02-26 |
| Examining board members: |
Jeniffer de Cassia Rillo Dutka;
Ana Paula Fukushiro;
Viviane Cristina de Castro Marino
|
| Advisor: | Jeniffer de Cassia Rillo Dutka |
| Abstract | |
Objectives: to establish anchor samples using speech recording rated with consensus regarding presence or absence of glottal (GS) and to compare auditory-perceptual ratings of GS with and without use of the anchor samples. Methods: the study was divided into two phases. During the first phase 480 phrases with stop and fricative sounds produced by speakers with history of cleft lip and palate were rated by three speech-language-pathologists (SLP) experienced with identification of GS. The phrases were rated individually by the three judges and those without consensus regarding the GS were rated again simultaneously. The samples rated with consensus regarding the presence or the absence of the GS were selected to establish a Bank of Anchor Samples Representative of GS during production of the 6 stops and 6 fricatives consonant targets. The second phase involved the selection of 48 anchor samples representing the 12 consonants of interest in this study and 120 experimental samples, along with the ratings of the experimental samples by three groups, each with three judges with distinct experience regarding the perceptual assessment of Cleft Palate Speech. The judges rated the experimental samples twice, first without the anchor samples and a week later accessing the anchor samples. Results: the rating establish during phase 1 indicated consensus regarding the GS for 352 samples, 120 representative of adequate production of the sounds of interest and 232 representative of use of GS. These 352 samples were used to establish the Bank of Anchor Samples Representative of GS. The results of the second phase revealed that while comparing ratings with and without access to anchor samples using mean Kappa Statistics, the interrater agreement for group 1 (G1) improved from regular (K=0,35) to moderate (K=0,55), for group 2 (G2) improved from moderate (K=0,44) to substantial (K=0,76) and for group 3 (G3) improved from substantial (K=0,72) to almost perfect (K=0,83). The best agreement was observed among the judges with most experience (G3) followed by recently graduated SLP (G2) and by the least experienced SLP students (G1). Conclusion: a Bank of Anchor Samples Representative of GS was established and the auditory perceptual ratings established after accessing the anchor samples were obtained with better interrater agreement and higher percentages of correct answers. The results suggest the importance of the use of anchor samples to minimize subjectivity of auditory perceptual evaluation of speech. (AU) | |
| FAPESP's process: | 14/07586-3 - Influence of use of anchor samples during auditory perceptual ratings of compensatory articulation in cleft lip and palate |
| Grantee: | Diana Conceição da Rocha Cardoso |
| Support Opportunities: | Scholarships in Brazil - Master |