An alaryngeal speech enhancement method based on ADPCM approach

An esophageal speech enhancement system is proposed to improve the intelligibility and quality of esophageal speech. Proposed system identifies the voiced segments of esophageal speech signal, by using pattern recognition methods, and replaces these by their equivalent voiced segments of normal speech, while keeps the silence and unvoiced segments without change. Evaluation results, using objective and subjective evaluation methods, show that proposed system provides a fairly good improvement of the quality and intelligibility of, esophageal speech signals.