日本バーチャルリアリティ学会論文誌
Online ISSN : 2423-9593
Print ISSN : 1344-011X
ISSN-L : 1344-011X
透過型HMDと音声認識技術を利用した聴覚補助方式の提案(<特集>福祉とVR)
井野 秀一小坂井 敦奈良 博之服部 裕之伊福部 達
著者情報
ジャーナル フリー

2001 年 6 巻 3 号 p. 177-183

詳細
抄録
There are still some problems regarding the recognition ability ofa speakerindependent speech recognition system. However, by using it as an interface between humans, they are able to understand the meaning of sentences, even with incorrect words. In this study, we have investigated how the understanding of sentences changes by adding nonverbal information, including a speaker's face, gestures and lip movements. In the experiment, we displayed letters consisting of incomplete sentences produced from a speech recognition system, together with a speaker's face, to subjects using a see-through HMD. Our results suggest that a hearing aid system which uses automatic speech recognition combined with a see-through HMD improves comprehension.
著者関連情報
© 2001 特定非営利活動法人 日本バーチャルリアリティ学会
前の記事 次の記事
feedback
Top