A whole word recurrent neural network for keyword spotting
- 1 January 1992
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 81-84 vol.2
- https://doi.org/10.1109/icassp.1992.226115
Abstract
The authors present a neural network which is trained on word examples to perform the wordspotting task. This network has multiple recurrent connections with time delay to account for temporal dynamics. A single network may be trained to recognize one word or many words. A hybrid wordspotter is evaluated in which a conventional wordspotter (based on dynamic time warping word matching) is used to screen incoming speech for potential keywords which are then passed to the network for the final accept/reject decision. Initial tests on a standard wordspotting test corpora resulted in improved keyword recognition at false alarm rates above zero.Keywords
This publication has 4 references indexed in Scilit:
- Robust syntax free speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A hidden Markov model based keyword recognition systemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Representation and Structure in Connectionist ModelsPublished by Defense Technical Information Center (DTIC) ,1989
- Complete gradient optimization of a recurrent network applied to /b/, /d/, /g/ discriminationThe Journal of the Acoustical Society of America, 1988