Visualization of Voice Disorders Using the Sammon Transform
The Sammon Transform performs data projections in a topology-preserving manner on the basis of an arbitrary distance measure. We use the weights of the observation probabilities of semi-continuous HMMs that were adapted to the current speaker as input. Experiments on laryngectomized speakers with tracheoesophageal substitute voice, hoarse, and normal speakers show encouraging results. Different speaker groups are separated in 2-D space, and the projection of a new speaker into the Sammon map allows prediction of his or her kind of voice pathology. The method can thus be used as an objective, automated support for the evaluation of voice disorders, and it visualizes them in a way that is convenient for speech therapists.
KeywordsSpeech Recognition Speech Recognition System Speech Effort Voice Disorder Speech Recognizer
Unable to display preview. Download preview PDF.
- 3.Steidl, S., Stemmer, G., Hacker, C., Nöth, E.: Adaption in the Pronunciation Space for Non-Native Speech Recognition. In: Proc. ICSLP, Jeju Island, Korea, pp. 318–321 (2004)Google Scholar
- 4.Shozakai, M., Nagino, G.: Analysis of Speaking Styles by Two-Dimensional Visualization of Aggregate of Acoustic Models. In: Proc. ICSLP, Jeju Island, Korea, pp. 717–720 (2004)Google Scholar
- 5.Schuster, M., Nöth, E., Haderlein, T., Steidl, S., Batliner, A., Rosanowski, F.: Can You Understand Him? Let’s Look at His Word Accuracy – Automatic Evaluation of Tracheoesophageal Speech. In: Proc. ICASSP, Philadelphia, PA, vol. I, pp. 61–64 (2005)Google Scholar