Speech Interface Exploiting Nonverbal Information

This paper introduces our research on speech interfaces using nonverbal information and examines new possibilities in speech interfaces. Although speech information consists of verbal and nonverbal information, most speech-recognition research has made use of only verbal information. From among nonverbal information, we have focused on hesitation (filled pause) and prosody (voice pitch) to create four speech-interface functions: Speech Completion, Speech Shift, Speech Starter, and Speech Spotter. These functions are based on simple, easy-to-utter rules and can be used without the need for training. By having users intentionally utter nonverbal information according to those rules, we have achieved interfaces that can exploit the potential of speech in various forms.