===== TDNN-Extracted features =====

//Abstract//: Time Delay Neural Network (TDNN) is a technique, derived from MLP, which performs a time invariant
processing in its lowest layers. This time invariant processing may be extracted from the network, in order
to code the speech for an other classifier such as Dynamic Time Warping (DTW). The resulting hybrid
system shows improved performances, with respect to both techniques used in isolation.
This paper describes this technique, gives results on a multi-speaker, isolated word recognition task, and
discusses its advantages.


<box 99% orange>
Xavier Driancourt and Léon Bottou:  **TDNN-Extracted features**,  //Proceedings of Neuro Nimes 90//, EC2, Nimes, France, 1990.

[[http://leon.bottou.org/publications/djvu/nimes-1990.djvu|nimes-1990.djvu]]
[[http://leon.bottou.org/publications/pdf/nimes-1990.pdf|nimes-1990.pdf]]
[[http://leon.bottou.org/publications/psgz/nimes-1990.ps.gz|nimes-1990.ps.gz]]
</box>

  @inproceedings{driancourt-bottou-90,
    author = {Driancourt, Xavier and Bottou, {L\'eon}},
    title = {{TDNN}-Extracted features},
    booktitle = {Proceedings of Neuro Nimes 90},
    publisher = {EC2},
    address = {Nimes, France},
    year = {1990},
    url = {http://leon.bottou.org/papers/driancourt-bottou-90},
  }