RECOGNIZING OF MIXED SPEECH
作者： Yu Dong , Weng Chao , Seltzer Michael L , Droppo James
DOI:
关键词:
摘要: FIELD: data processing.SUBSTANCE: invention relates to means of recognizing mixed speech. First neural network is trained recognize a speech signal pronounced by speaker with higher level characteristics from sample Second lower Mixed decoded first and second optimizing the combined probability observing said two signals, where that particular frame switching point characteristic. Third taught predict based on prediction.EFFECT: high accuracy speech.15 cl, 5 tbl, 6 dwg
lens.org 参考文章(8)
Paris Smaragdis, Systems and Methods for Concurrent Signal Recognition ,(2011)
Mark Hanegraaff, Jamey Poirier, Darrell Poirier, Enhancement of simultaneous multi-user real-time speech recognition system ,(2007)
Julian Odell, Jon Hamaker, Nikko Strom, Progressive application of knowledge sources in multistage speech recognition ,(2009)
Franz Gerl, Tobias Herbig, Speaker Recognition in a Speech Recognition System ,(2010)
Tadashi Emori, Masahiro Tani, Yoshifumi Onishi, Speaker selecting device, speaker adaptive model creating device, speaker selecting method, speaker selecting program, and speaker adaptive model making program ,(2008)
Shinya Matsui, Katsumasa Nagahama, Sound source separating device, speech recognizing device, portable telephone, and sound source separating method, and program ,(2006)
Andrianov, Bochkarev V L, Os Kin V A, Ivanov A I, Bochkarev S L, AUTOMATIC PERSON IDENTIFICATION METHOD ,(2001)