Robustness analysis on lattice-based speech indexing approaches with to varying recognition accuracies by refined simulations
Journal
2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings
Pages
289-292
Date Issued
2008
Author(s)
Abstract
We analyze the robustness of different lattice-based speech indexing approaches. While we believe such analysis is important, to our knowledge it has been neglected in prior works. In order to make up for the lack of corpora with various noise characteristics, we use refined approaches to simulate feature vector sequences directly from HMMs, including those with a wide range of recognition accuracies, as opposed to simply adding noise and channel distortion to the existing noisy corpora. We compare, analyze, and discuss the robustness of several state-of-the-art speech indexing approaches.
SDGs
Type
conference paper