%0 Journal Article %1 waibel1989phoneme %A Waibel, A. %A Hanazawa, T. %A Hinton, G. %A Shikano, K. %A Lang, K. J. %D 1989 %I IEEE %J IEEE Transactions on Acoustics, Speech, and Signal Processing %K chm1320 %N 3 %P 328--339 %R 10.1109/29.21701 %T Phoneme recognition using time-delay neural networks %U http://dx.doi.org/10.1109/29.21701 %V 37 %X The authors present a time-delay neural network (TDNN) approach to phoneme recognition which is characterized by two important properties: (1) using a three-layer arrangement of simple computing units, a hierarchy can be constructed that allows for the formation of arbitrary nonlinear decision surfaces, which the TDNN learns automatically using error backpropagation; and (2) the time-delay arrangement enables the network to discover acoustic-phonetic features and the temporal relationships between them independently of position in time and therefore not blurred by temporal shifts in the input. As a recognition task, the speaker-dependent recognition of the phonemes B, D, and G in varying phonetic contexts was chosen. For comparison, several discrete hidden Markov models (HMM) were trained to perform the same task. Performance evaluation over 1946 testing tokens from three speakers showed that the TDNN achieves a recognition rate of 98.5\% correct while the rate obtained by the best of the HMMs was only 93.7\%

@article{waibel1989phoneme, abstract = {{The authors present a time-delay neural network (TDNN) approach to phoneme recognition which is characterized by two important properties: (1) using a three-layer arrangement of simple computing units, a hierarchy can be constructed that allows for the formation of arbitrary nonlinear decision surfaces, which the TDNN learns automatically using error backpropagation; and (2) the time-delay arrangement enables the network to discover acoustic-phonetic features and the temporal relationships between them independently of position in time and therefore not blurred by temporal shifts in the input. As a recognition task, the speaker-dependent recognition of the phonemes B, D, and G in varying phonetic contexts was chosen. For comparison, several discrete hidden Markov models (HMM) were trained to perform the same task. Performance evaluation over 1946 testing tokens from three speakers showed that the TDNN achieves a recognition rate of 98.5\% correct while the rate obtained by the best of the HMMs was only 93.7\%}}, added-at = {2017-07-19T15:29:59.000+0200}, author = {Waibel, A. and Hanazawa, T. and Hinton, G. and Shikano, K. and Lang, K. J.}, biburl = {https://www.bibsonomy.org/bibtex/2f7d66e9f91f5fda316b4bf1687f522f0/andreashdez}, citeulike-article-id = {1290889}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/29.21701}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs\_all.jsp?arnumber=21701}, day = 06, doi = {10.1109/29.21701}, institution = {Dept. of Comput. Sci., Carnegie-Mellon Univ., Pittsburgh, PA, USA}, interhash = {2bb0adcf7995d4357205e6a723df20bf}, intrahash = {f7d66e9f91f5fda316b4bf1687f522f0}, issn = {00963518}, journal = {IEEE Transactions on Acoustics, Speech, and Signal Processing}, keywords = {chm1320}, month = mar, number = 3, pages = {328--339}, posted-at = {2017-05-21 18:28:06}, priority = {2}, publisher = {IEEE}, timestamp = {2017-07-19T15:31:02.000+0200}, title = {{Phoneme recognition using time-delay neural networks}}, url = {http://dx.doi.org/10.1109/29.21701}, volume = 37, year = 1989 }