Untitled

Finan, Robert; Damper, R. I.; Sapeluk, A.T.

doi:10.1023/a:1009652732313

Cited by 8 publications

(1 citation statement)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent attention in speaker recognition has focussed on the use of subband processing, whereby the wideband signal is fed to a bank of bandpass filters to give a set of time-varying outputs, which are individually processed before using multiple classifier techniques to produce a combined, overall decision [2, 3, 22, 131, see Figure 1. Because the subband signals vary slowly relative to the wideband signal, the problem of representing them by some data model should be simplified [9]. Previous work has demonstrated the performance gains that can be obtained in speaker recognition by applying subband processing, together with hidden Markov modelling and multiple classifier recombination.…”

Section: Introductionmentioning

confidence: 99%

Information fusion for subband-HMM speaker recognition

Higgins

Damper

Dodd

IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222)

Self Cite

View full text Add to dashboard Cite

Previous work has demonstrated the performance gains that can be obtained in speaker recognition by applying subband processing, together with hidden Markov modelling and multiple classifier recombination. Two recombination rules have been investigated: the sum of log likelihoods, which corresponds to the optimal Bayes' rule under certain constraints, and multilayer perceptrons (MLP), which are not subject to these constraints. It was found that for two spoken digits in the presence of a single case of narrowband noise the sum of log likelihoods and MLP achieved comparable performance. In this paper, the previous work is extended in the direction of investigating the robustness of the recognition system to different narrowband noise. Two approaches are taken towards this aim. Firstly, narrowband noise is added at different centre frequencies. Secondly, a Bayesian MLP approach is investigated using automatic relevance determination (ARD) on the subband inputs to the MLP. From this it is possible to assess the relative importance of the subbands to recognition performance. Results for the new noise conditions show that the sum of log likelihoods generally does better than the (average) MLP fusion.

show abstract

Section: Introductionmentioning

confidence: 99%