単語認識と連続音節認識を併用した音声認識とタッチ入力を利用した姓名音声入力方式

概要

論文の詳細を見る
We propose an interface for a name input based on speech recognition using a word dictionary and syllable-based N-gram. User first utters a name and then chooses the correct word/syllables by touch input from word/syllable candidates which were obtained from speech recognition. Name utterance is hard to recognize accurately because of the large vocabulary size, so the system uses isolated word recognition with a dictionary containing high frequent words and continuous syllable recognition with syllable-based N-gram. System displays word candidates, syllable sequence candidates and a syllable lattice on the touch panel. Users who were familiar to speech input could find the correct answer from word candidates or syllable sequence candidates at a rate of 87-92%, and they felt that it was useful. Novice users also obtain the same level of accurate candidates after 30-40 minutes' practice and the interface became useful to them.