観音システム 有音/無音, 有声音/無声音, 母音の認識

中室 健; 春木 克洋; 堀口 竜平; 中野 沙路麿; 杉本 末雄

doi:10.11509/sci.SCI06.0.166.0

The 50th Annual Conference of the Institute of Systems, Control and Information Engineers

Session ID : 5F1-3

DOI https://doi.org/10.11509/sci.SCI06.0.166.0

Conference information

Host: The Institute of Systems, Control and Information Engineers

The KanNon System

speech/silence, voiced/unvoiced and vowel recognition

ken nakamuro, katsuhiro haruki, *ryuhei horiguchi, saroma nakano, sueo sugimoto

Author information

Keywords: speech visualization, neural network, speech/silence recognition, voiced/unvoiced recognition, vowel recognition

CONFERENCE PROCEEDINGS FREE ACCESS

Details

Abstract

We study about the speech visualization system "KanNon" which supports deaf people's life. To realize speech visualization, the real-time phoneme and syllable level speech recognition system is needed. So we developed the speech recognition systems for speech/silence, voiced/unvoiced and Japanese 5 vowels recognition using neural networks. Finally we show experimental results of our proposed system.

Corresponding author

Register with J-STAGE for free!