Language selection

Search


Neural network classifier architectures for phoneme recognition / by William Treurniet.Co24-3/8-1992-1E-PDF

"Automatic recognition of words in continuous speech is difficult to do with whole-word template models, especially when the vocabulary is reasonably large. Instead, the currently preferred approaches for this task hypothesize the presence of words or sub-words, such as syllables or phonemes, on the basis of likelihood estimates obtained from comparisons of the acoustic data with statistical models derived from training data. This paper is concerned with the application of artificial neural networks, trained with the back-propagation learning algorithm, to modelling phonemes extracted from the DARPA TIMIT multi-speaker, continuous speech data base"--Abstract.

Permanent link to this Catalogue record:
publications.gc.ca/pub?id=9.890766&sl=0

Publication information
Department/Agency
  • Communications Research Centre (Canada), issuing body.
TitleNeural network classifier architectures for phoneme recognition / by William Treurniet.
Series title
  • CRC technical note ; no. CRC-TN-92-001
Publication typeMonograph - View Master Record
Language[English]
FormatDigital text
Electronic document
Note(s)
  • "Ottawa, March 26, 1992."
  • Digitized edition from print [produced by Innovation, Science and Economic Development Canada].
  • Includes bibliographical references (pages 36-38).
  • Issued also in print format.
  • Includes abstracts in English and French.
Publishing information
  • Ottawa : Communications Research Centre, Broadcast Technologies Research = Centre de recherches sur les communications, Recherche en technologies de radiodiffusion, 1992.
  • ©1992.
Author / Contributor
  • Treurniet, W. C. (William C.)author.
Description1 online resource (iii, 38 pages) : illustrations.
Catalogue number
  • Co24-3/8-1992-1E-PDF
Subject terms
Request alternate formats
To request an alternate format of a publication, complete the Government of Canada Publications email form. Use the form’s “question or comment” field to specify the requested publication.

Page details