8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Using Acoustic Models to Choose Pronunciation Variations for Synthetic Voices

Christina L. Bennett, Alan W. Black

Carnegie Mellon University, USA

Within-speaker pronunciation variation is a well-known phenomenon; however, attempting to capture and predict a speaker's choice of pronunciations has been mostly overlooked in the field of speech synthesis. We propose a method to utilize acoustic modeling techniques from speech recognition in order to detect a speaker's choice between full and reduced pronunciations.

Full Paper

Bibliographic reference.  Bennett, Christina L. / Black, Alan W. (2003): "Using acoustic models to choose pronunciation variations for synthetic voices", In EUROSPEECH-2003, 2937-2940.