TY - JOUR
T1 - Ears of the robot
T2 - Three simultaneous speech segregation and recognition using robot-mounted microphones
AU - Mochiki, Naoya
AU - Ogawa, Tetsuji
AU - Kobayashi, Tetsunori
PY - 2007/9
Y1 - 2007/9
N2 - A new type of sound source segregation method using robot-mounted microphones, which are free from strict head related transfer function (HRTF) estimation, has been proposed and successfully applied to three simultaneous speech recognition systems. The proposed segregation method is executed with sound intensity differences that are due to the particular arrangement of the four directivity microphones and the existence of a robot head acting as a sound barrier. The proposed method consists of three-layered signal processing: two-line SAFIA (binary masking based on the narrow band sound intensity comparison), two-line spectral subtraction and their integration. We performed 20 K vocabulary continuous speech recognition test in the presence of three speakers' simultaneous talk, and achieved more than 70% word error reduction compared with the case without any segregation processing.
AB - A new type of sound source segregation method using robot-mounted microphones, which are free from strict head related transfer function (HRTF) estimation, has been proposed and successfully applied to three simultaneous speech recognition systems. The proposed segregation method is executed with sound intensity differences that are due to the particular arrangement of the four directivity microphones and the existence of a robot head acting as a sound barrier. The proposed method consists of three-layered signal processing: two-line SAFIA (binary masking based on the narrow band sound intensity comparison), two-line spectral subtraction and their integration. We performed 20 K vocabulary continuous speech recognition test in the presence of three speakers' simultaneous talk, and achieved more than 70% word error reduction compared with the case without any segregation processing.
KW - Robot audition
KW - SAFIA
KW - Sound source segregation
KW - Spectral subtraction
KW - Speech recognition
UR - http://www.scopus.com/inward/record.url?scp=68249135534&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=68249135534&partnerID=8YFLogxK
U2 - 10.1093/ietisy/e90-d.9.1465
DO - 10.1093/ietisy/e90-d.9.1465
M3 - Article
AN - SCOPUS:68249135534
SN - 0916-8532
VL - E90-D
SP - 1465
EP - 1468
JO - IEICE Transactions on Information and Systems
JF - IEICE Transactions on Information and Systems
IS - 9
ER -