TY - GEN
T1 - Eyes free dialing for cellular telephones
AU - Bendelac, G.
AU - Shallom, I. D.
AU - Markovitch, I.
N1 - Publisher Copyright:
© Proceedings - 17th Convention of Electrical and Electronics Engineers in Israel, EEIS 1991. All rights reserved.
PY - 1991/1/1
Y1 - 1991/1/1
N2 - The widespread use of cellular telephones in handsfree mode, has created a need for high performance speech recognition systems for voice dialing. This paper describes a chip set, the DSPG5006, that implements digital signal processing algorithms for voice dialing in cellular telephones. The problem of achieving high performance speech recognition in noisy car cabins is addressed by using noise immune techniques for preprocessing, feature extraction, template training and template comparison. The chip set supports an isolated word, speaker dependent, speech recognition system that is built around a modified dynamic time warping algorithm. A speech specific voice activated switch is used in conjunction with a simplified endpoint detector to reliably define the word boundaries. A two stage training mode permits building of compact and reliable templates that are compared to incoming words in the recognition mode. Recognition rates are greater than 98% in car cabins, and over 99% in the laboratory. The chip set can be configured to support speech synthesis, so that the user is guided in a friendly manner throughout the use of the system, for both"prompting and verification. It is possible to use the chip set for dialing digit by digit or more effectively, by using user defined phrases, in a similar way to dialing memory stored numbers in digital phones. The system can support a number of users, depending on the memory size of the application.
AB - The widespread use of cellular telephones in handsfree mode, has created a need for high performance speech recognition systems for voice dialing. This paper describes a chip set, the DSPG5006, that implements digital signal processing algorithms for voice dialing in cellular telephones. The problem of achieving high performance speech recognition in noisy car cabins is addressed by using noise immune techniques for preprocessing, feature extraction, template training and template comparison. The chip set supports an isolated word, speaker dependent, speech recognition system that is built around a modified dynamic time warping algorithm. A speech specific voice activated switch is used in conjunction with a simplified endpoint detector to reliably define the word boundaries. A two stage training mode permits building of compact and reliable templates that are compared to incoming words in the recognition mode. Recognition rates are greater than 98% in car cabins, and over 99% in the laboratory. The chip set can be configured to support speech synthesis, so that the user is guided in a friendly manner throughout the use of the system, for both"prompting and verification. It is possible to use the chip set for dialing digit by digit or more effectively, by using user defined phrases, in a similar way to dialing memory stored numbers in digital phones. The system can support a number of users, depending on the memory size of the application.
UR - http://www.scopus.com/inward/record.url?scp=84991784673&partnerID=8YFLogxK
U2 - 10.1109/EEIS.1991.217655
DO - 10.1109/EEIS.1991.217655
M3 - Conference contribution
AN - SCOPUS:84991784673
T3 - Proceedings - 17th Convention of Electrical and Electronics Engineers in Israel, EEIS 1991
SP - 234
EP - 237
BT - Proceedings - 17th Convention of Electrical and Electronics Engineers in Israel, EEIS 1991
PB - Institute of Electrical and Electronics Engineers
T2 - 17th Convention of Electrical and Electronics Engineers in Israel, EEIS 1991
Y2 - 5 March 1991 through 7 March 1991
ER -