Isolated Digit Recognition with LSTM Recurrent Networks. Graves, A., Eck, D., Beringer, N., & Schmidhuber, J. In First International Workshop on Biologically Inspired Approaches to Advanced Information Technology, Lausanne, 2004. abstract bibtex Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) are local in space and time and closely related to a biological model of memory in the prefrontal cortex. Not only are they more biologically plausible than previous artificial RNNs, they also outperformed them on many artificially generated sequential processing tasks. This encouraged us to apply LSTM to more realistic problems, such as the recognition of spoken digits. Without any modification of the underlying algorithm, we achieved results comparable to state-of-the-art Hidden Markov Model (HMM) based recognisers on both the TIDIGITS and TI46 speech corpora. We conclude that LSTM should be further investigated as a biologically plausible basis for a bottom-up, neural net-based approach to speech recognition.
@INPROCEEDINGS{graves+eck+beringer+schmidhuber:2004,
AUTHOR = {A. Graves and D. Eck and N. Beringer and J. Schmidhuber},
TITLE = {Isolated Digit Recognition with {LSTM} Recurrent Networks},
BOOKTITLE = {First International Workshop on Biologically Inspired Approaches to Advanced Information Technology},
ADDRESS = {Lausanne},
YEAR = {2004},
SOURCE = {OwnPublication},
ABSTRACT = {Long Short-Term Memory (LSTM) recurrent neural networks
(RNNs) are local in space and time and closely related to a biological model of memory in the
prefrontal cortex. Not only are they more biologically plausible than previous artificial RNNs,
they also outperformed them on many artificially generated sequential processing tasks.
This encouraged us to apply LSTM to more realistic problems, such as the recognition of spoken digits.
Without any modification of the underlying algorithm, we achieved results comparable to state-of-the-art Hidden Markov Model
(HMM) based recognisers on both the TIDIGITS and TI46 speech corpora.
We conclude that LSTM should be further investigated as a biologically plausible basis for a bottom-up,
neural net-based approach to speech recognition.}
}
Downloads: 0
{"_id":"ZFZpPvejYMzjK5mMi","bibbaseid":"graves-eck-beringer-schmidhuber-isolateddigitrecognitionwithlstmrecurrentnetworks-2004","downloads":0,"creationDate":"2018-10-05T11:34:49.675Z","title":"Isolated Digit Recognition with LSTM Recurrent Networks","author_short":["Graves, A.","Eck, D.","Beringer, N.","Schmidhuber, J."],"year":2004,"bibtype":"inproceedings","biburl":"http://people.idsia.ch/~juergen/deep.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["A."],"propositions":[],"lastnames":["Graves"],"suffixes":[]},{"firstnames":["D."],"propositions":[],"lastnames":["Eck"],"suffixes":[]},{"firstnames":["N."],"propositions":[],"lastnames":["Beringer"],"suffixes":[]},{"firstnames":["J."],"propositions":[],"lastnames":["Schmidhuber"],"suffixes":[]}],"title":"Isolated Digit Recognition with LSTM Recurrent Networks","booktitle":"First International Workshop on Biologically Inspired Approaches to Advanced Information Technology","address":"Lausanne","year":"2004","source":"OwnPublication","abstract":"Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) are local in space and time and closely related to a biological model of memory in the prefrontal cortex. Not only are they more biologically plausible than previous artificial RNNs, they also outperformed them on many artificially generated sequential processing tasks. This encouraged us to apply LSTM to more realistic problems, such as the recognition of spoken digits. Without any modification of the underlying algorithm, we achieved results comparable to state-of-the-art Hidden Markov Model (HMM) based recognisers on both the TIDIGITS and TI46 speech corpora. We conclude that LSTM should be further investigated as a biologically plausible basis for a bottom-up, neural net-based approach to speech recognition.","bibtex":"@INPROCEEDINGS{graves+eck+beringer+schmidhuber:2004,\n AUTHOR = {A. Graves and D. Eck and N. Beringer and J. Schmidhuber},\n TITLE = {Isolated Digit Recognition with {LSTM} Recurrent Networks},\n BOOKTITLE = {First International Workshop on Biologically Inspired Approaches to Advanced Information Technology},\n ADDRESS = {Lausanne},\n YEAR = {2004},\n SOURCE = {OwnPublication},\n ABSTRACT = {Long Short-Term Memory (LSTM) recurrent neural networks \n\t\t (RNNs) are local in space and time and closely related to a biological model of memory in the \n\t\t prefrontal cortex. Not only are they more biologically plausible than previous artificial RNNs, \n\t\t they also outperformed them on many artificially generated sequential processing tasks. \n\t\t This encouraged us to apply LSTM to more realistic problems, such as the recognition of spoken digits. \n\t\t Without any modification of the underlying algorithm, we achieved results comparable to state-of-the-art Hidden Markov Model \n\t\t (HMM) based recognisers on both the TIDIGITS and TI46 speech corpora. \n\t\t We conclude that LSTM should be further investigated as a biologically plausible basis for a bottom-up, \n\t\t neural net-based approach to speech recognition.}\n}\n\n","author_short":["Graves, A.","Eck, D.","Beringer, N.","Schmidhuber, J."],"key":"graves+eck+beringer+schmidhuber:2004","id":"graves+eck+beringer+schmidhuber:2004","bibbaseid":"graves-eck-beringer-schmidhuber-isolateddigitrecognitionwithlstmrecurrentnetworks-2004","role":"author","urls":{},"downloads":0,"html":""},"search_terms":["isolated","digit","recognition","lstm","recurrent","networks","graves","eck","beringer","schmidhuber"],"keywords":[],"authorIDs":[],"dataSources":["EmYaiv9TCHbg7caTW"]}