Speech recognition on an FPGA using discrete and continuous hidden Markov models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

Speech recognition on an FPGA using discrete and continuous hidden Markov models. / Melnikoff, Stephen; Quigley, Steven; Russell, Martin.

Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings. Springer, 2002. p. 202-211 (Lecture Notes in Computer Science; Vol. 2438).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Melnikoff, S, Quigley, S & Russell, M 2002, Speech recognition on an FPGA using discrete and continuous hidden Markov models. in Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings. Lecture Notes in Computer Science, vol. 2438, Springer, pp. 202-211, 12th International Conference on Field-Programmable Logic and Applications, Sep 02-04, 2002. MONTPELLIER, France, 1/01/02. <http://www.springerlink.com/content/12lmfc9httabcrae>

APA

Melnikoff, S., Quigley, S., & Russell, M. (2002). Speech recognition on an FPGA using discrete and continuous hidden Markov models. In Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings (pp. 202-211). (Lecture Notes in Computer Science; Vol. 2438). Springer. http://www.springerlink.com/content/12lmfc9httabcrae

Vancouver

Melnikoff S, Quigley S, Russell M. Speech recognition on an FPGA using discrete and continuous hidden Markov models. In Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings. Springer. 2002. p. 202-211. (Lecture Notes in Computer Science).

Author

Melnikoff, Stephen ; Quigley, Steven ; Russell, Martin. / Speech recognition on an FPGA using discrete and continuous hidden Markov models. Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings. Springer, 2002. pp. 202-211 (Lecture Notes in Computer Science).

Bibtex

@inproceedings{171d18beee9c4ae28f0d05c57c8e0806,
title = "Speech recognition on an FPGA using discrete and continuous hidden Markov models",
abstract = "Speech recognition is a computationally demanding task, particularly the stage which uses Viterbi decoding for converting pre-processed speech data into words or sub-word units. Any device that can reduce the load on, for example, a PC{\textquoteright}s processor, is advantageous. Hence we present FPGA implementations of the decoder based alternately on discrete and continuous hidden Markov models (HMMs) representing monophones, and demonstrate that the discrete version can process speech nearly 5,000 times real time, using just 12% of the slices of a Xilinx Virtex XCV1000, but with a lower recognition rate than the continuous implementation, which is 75 times faster than real time, and occupies 45% of the same device.",
author = "Stephen Melnikoff and Steven Quigley and Martin Russell",
note = "The original publication is available at www.springerlink.com; 12th International Conference on Field-Programmable Logic and Applications, Sep 02-04, 2002. MONTPELLIER, France ; Conference date: 01-01-2002",
year = "2002",
month = jan,
day = "1",
language = "English",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "202--211",
booktitle = "Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings",

}

RIS

TY - GEN

T1 - Speech recognition on an FPGA using discrete and continuous hidden Markov models

AU - Melnikoff, Stephen

AU - Quigley, Steven

AU - Russell, Martin

N1 - The original publication is available at www.springerlink.com

PY - 2002/1/1

Y1 - 2002/1/1

N2 - Speech recognition is a computationally demanding task, particularly the stage which uses Viterbi decoding for converting pre-processed speech data into words or sub-word units. Any device that can reduce the load on, for example, a PC’s processor, is advantageous. Hence we present FPGA implementations of the decoder based alternately on discrete and continuous hidden Markov models (HMMs) representing monophones, and demonstrate that the discrete version can process speech nearly 5,000 times real time, using just 12% of the slices of a Xilinx Virtex XCV1000, but with a lower recognition rate than the continuous implementation, which is 75 times faster than real time, and occupies 45% of the same device.

AB - Speech recognition is a computationally demanding task, particularly the stage which uses Viterbi decoding for converting pre-processed speech data into words or sub-word units. Any device that can reduce the load on, for example, a PC’s processor, is advantageous. Hence we present FPGA implementations of the decoder based alternately on discrete and continuous hidden Markov models (HMMs) representing monophones, and demonstrate that the discrete version can process speech nearly 5,000 times real time, using just 12% of the slices of a Xilinx Virtex XCV1000, but with a lower recognition rate than the continuous implementation, which is 75 times faster than real time, and occupies 45% of the same device.

M3 - Conference contribution

T3 - Lecture Notes in Computer Science

SP - 202

EP - 211

BT - Field-Programmable Logic and Applications. Reconfigurable Computing Is Going Mainstream 12th International Conference, FPL 2002, Montpellier, France September 2-4, 2002. Proceedings

PB - Springer

T2 - 12th International Conference on Field-Programmable Logic and Applications, Sep 02-04, 2002. MONTPELLIER, France

Y2 - 1 January 2002

ER -