Audiovusual automatic speech segmentation

Audiovisual speech segmentation using visual information together with audio data is introduced. The collaboration of audio and visual data results in lower average absolute boundary error between the manual segmentation and automatic segmentation results that directly affects the quality of speech processing systems using the segmented database. The audio and visual feature vectors are fused at the feature level and used in a HMM based speech segmentation system. A Turkish audiovisual speech database has been prepared and used in the experiments. The average absolute boundary error decreases up to 20.82% by using different audiovisual feature vectors.