Impact of auditory training for perceptual assessment of voice executed by undergraduate students in Speech-Language Pathology. References Publications referenced by this paper. Topics Discussed in This Paper. Skip to search form Skip to main content. Assessment of vocal cord nodules: Voice analysis of postlingually deaf adults pre- and postcochlear implantation. See our FAQ for additional information.
|Published (Last):||27 June 2007|
|PDF File Size:||3.67 Mb|
|ePub File Size:||8.21 Mb|
|Price:||Free* [*Free Regsitration Required]|
Introduction Voice disorders may be caused by voice-strain due to speaking or singing, vocal cord damage, infection, side-effects of inhaled steroids as used to treat asthma or more serious disease including laryngeal cancer.
The resulting loss of voice quality can be measured subjectively or objectively. The GRBAS scale has the advantage of being widely understood and recommended by many professional bodies, but its subjectivity and reliance on highly trained personnel are significant limitations. The aim of this research is to design and evaluate objective measurement of voice quality conforming to the GRBAS standard. Figure 1. Overview of the project 2. Voice Collection About voice recordings from patients and volunteers were collected.
Figure 2. GRBAS presentation and scoring package 2. Figure 3. Agreement between SLTs 3. This classifier was designed to distinguish, firstly, samples scoring 0 from those scoring 1, 2 or 3, secondly, samples scoring 1 from those scoring 2 or 3, and finally, samples scoring 2 from those scoring 3. Figure 4, shows the average of precision, recall and F-measure. Figure 4.
Average of Precision, Recall, F-measure 3-stage classifier 4. To do objective analysis voice features were extracted by Kaypentax and they were used for training the classifier in predicting GRBAS scores. To improve the precision appropriate voice features should be investigated.
GRBAS SCALE PDF