Comparison of Parametric Spectral Representations for Voice Recognition in Noisy Environments
暂无分享,去创建一个
ABSTRACf Voice recognition systems provide good performances when the speech signal is recorded in good conditions: low noise level. good microphones. But results are not sufficient for several reaIlife noisy situations (e.g cars). The aim of the presented work was to compare three techniques of spectral parametrisation in terms of performances for speech recognition and more precisely to evaluate their robustness in noise. This study was part of a French GRECO project on the comparison of methods of parametric and non parametric spectral analysis for speech recognition. This project has used the existing speech recognition program SAMREC-l with the speech data base EUROMO and the RSG_I0 noise data base. The different techniques are evaluated by their scores of recognition in lexical accesses. for speaker dependent isolated word recognition based on Dynamic Time Warping.
[1] Fumitada Itakura,et al. Distance measure for speech recognition based on the smoothed group delay spectrum , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.
[2] Bayya Yegnanarayana,et al. Speech processing using group delay functions , 1991, Signal Process..
[3] K. K. Paliwal. A perception‐based LSP distance measure for speech recognition , 1988 .
[4] Sadaoki Furui,et al. Line spectrum pair frequency - based distance measures for speech recognition , 1990, ICSLP.