Hierarchical bottle neck features for LVCSR

  title={Hierarchical bottle neck features for LVCSR},
  author={Christian Plahl and Ralf Schl{\"u}ter and Hermann Ney},
This paper investigates the combination of different neural network topologies for probabilistic feature extraction. On one hand, a five-layer neural network used in bottle neck feature extraction allows to obtain arbitrary feature size without dimensionality reduction by transform, independently of the training targets. On the other hand, a hierarchical processing technique is effective and robust over several conditions. Even though the hierarchical and bottle neck processing performs equally… CONTINUE READING
Highly Cited
This paper has 47 citations. REVIEW CITATIONS


Publications referenced by this paper.
Showing 1-10 of 17 references

Similar Papers

Loading similar papers…