Speech completion is a novel speech interface function that helps a user enter a word or phrase by completing (filling in the rest of) a phrase fragment uttered by the user. Although the concept of completion is widely used in text-based interfaces, there have been no reports of completion being effectively applied to speech. By using a filled pause, we enable a user to effortlessly invoke the speech-completion function which helps the user recall uncertain phrases and saves labor when the input phrase is long. When a user hesitates by lengthening a vowel (a filled pause is uttered) during a phrase, our system immediately displays completion candidates whose beginnings acoustically resemble the uttered fragment so that the user can select the correct one. In our experiments with a system that included a filled-pause detector and a speech recognizer capable of listing candidates, the effectiveness of speech completion was confirmed.

Benefits Summary

The present invention relates to an apparatus, a method and a recording medium generally applied in the speech recognition. One reason that oral communication is an excellent means for human beings to exchange information is that a listener can help a speaker's speech act or concept-forming. In a human speech dialog, therefore, even when the speaker stumbles in his speech, the listener may guess what the speaker intends to say and suggest some candidates, thus helping the speaker remember or complete what he had intended to say. The concept of complementing has been widely applied to text interfaces. For example, several text editors (e.g., Emacs and Mule) and UNIX shells (e.g., tcsh and bash) provide the complementing function (called "completion") for file names and command names. In such a function, when the user presses a key (typically the Tab key) to call the complementing function (hereinafter referred to as "complementing trigger key"), the remaining portion of the fraction of a word that has been typed halfway is complemented. In WWW browsers such as Netscape Communication and Internet Explorer also, the automatic complementing function (called "autocompletion") for URLs has been introduced, wherein the system provides lists of complementing candidates one after another while the user is typing. Recently, the complementing function has been introduced into pen-based interfaces. For example, interfaces with automatic complementing functions such as a predictive pen-input interface and POBox have been proposed. For a speech-input interface, however, the speech complementing input has not been realized because until now there has been no appropriate means for calling the complementing function while the speech is being input. more

Development Summary

The technology can be demonstrated. Sample applications created during research include a speech-capable music jukebox system. This jukebox system can play back a song whose title is determined through speech recognition with the speech-completion function. The system has been demonstrated at Japanese exhibitions and received much publicity from the press. more

IP Summary

This technology is supported by 1 US patent. The most recent year of issue is 2005. more

Discussions (0 items)

No discussions have been created for this TechPak.