Tools for expressive text-to-speech markup

Abstract

This paper describes handicapped accessible text-to-speech markup software developed for poetry and performance. Most text-to-speech software allows the user to select a voice, but provides no control over performance parameters such as rate, volume, and pitch. For users with vocal disabilities, the default "computer voice" is often dreaded since it provides no personalization. Evolving standards exist for text-to-speech markup (Sable, Java Speech Markup Language, Spoken Text Markup Language), but few tools exist for non-experts to modify documents using these prosody options [1, 5]. Furthermore, we could find fewer tools allowing for straightforward live performance using a synthesized voice [3]. Thus we created an easy to learn text-to-speech markup tool that requires little training to use.