Time delay neural network: Difference between revisions

Revision as of 13:59, 10 May 2025

Revision as of 14:00, 10 May 2025

Line 1:

[[File:TDNN Diagram.png|thumb|right|TDNN diagram]]

'''Time delay neural network''' ('''TDNN''')<ref name="phoneme detection">[[Alex{{cite Waibel|Alexanderjournal Waibel]], Tashiyuki Hanazawa, [[Geoffrey Hinton]], Kiyohito Shikano, Kevin J|doi=10.1109/29.21701 Lang, ''[|url=http://www.inf.ufrgs.br/~engel/data/media/file/cmp121/waibel89_TDNN.pdf |title=Phoneme Recognitionrecognition Usingusing Timetime-Delaydelay Neuralneural Networks]'',networks |date=1989 |last1=Waibel |first1=A. |last2=Hanazawa |first2=T. |last3=Hinton |first3=G. |last4=Shikano |first4=K. |last5=Lang |first5=K.J. |journal=IEEE Transactions on Acoustics, Speech, and Signal Processing, Volume |volume=37, No. |issue=3, pp.|pages=328–339 328. - 339 March 1989.}}</ref> is a multilayer [[artificial neural network]] architecture whose purpose is to 1) classify patterns with shift-invariance, and 2) model context at each layer of the network.

Shift-invariant classification means that the classifier does not require explicit segmentation prior to classification. For the classification of a temporal pattern (such as speech), the TDNN thus avoids having to determine the beginning and end points of sounds before classifying them.