No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
The full text of this article is not currently available.
Temporal resolution analysis in frequency domain linear prediction
2. Bourlard, H. , and Morgan, N. (1994). Connectionist Speech Recognition—A Hybrid Approach (Kluwer Academic Publishers, MA).
3. Davis, S. , and Mermelstein, P. (1980). “Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences,” IEEE Trans. Acoust., Speech, Signal Process. 28, 357–366.
4. Ganapathy, S. , and Hermansky, H. (2012). “Robust phoneme recognition using high-resolution temporal envelopes,” in Proceedings of INTERSPEECH, Portland, OR.
5. Hermansky, H. (1983). “Analysis and synthesis of speech based on spectral transform linear predictive method,” in Proceedings of ICASSP, Boston, MA, Vol. 8, pp. 777–780.
6. Kumerasan, R. , and Rao, A. (1999). “Model-based approach to envelope and positive instantaneous frequency estimation of signals with speech applications,” J. Acoust. Soc. Am. 105(3 ), 1912–1924.
9. Pinto, J. , Yegnanarayana, B. , Hermansky, H. , and Doss, M. M. (2007). “Exploiting contextual information for improved phoneme recognition,” in Proceedings of INTERSPEECH, Antwerp, Belgium, pp. 1817–1820.
10. Rayleigh, L. (1880). “Scientific papers, Vol. 1,” Proc. London Math. Soc. 11, 57–80.
11. Reynolds, D. A. (1997). “HTIMIT and LLHDB: Speech corpora for the study of hand set transducer effects,” in Proceedings of ICASSP, Munich, pp. 1535–1538.
13. Thomas, S. , Ganapathy, S. , and Hermansky, H. (2008). “Recognition of reverberant speech using frequency domain linear prediction,” IEEE Signal Process. Lett. 15, 681–684.
Article metrics loading...
Frequency domain linear prediction (FDLP) is a technique for auto-regressive modeling of Hilbert envelopes. In this letter, the resolution properties of the FDLP model are investigated using synthetic signals with impulses immersed in noise. The effect of various factors are studied which affect the temporal resolution and this analysis suggests ways to improve the resolution of the FDLP envelopes in noisy speech. The high resolution FDLP envelopes are used to derive robust features for phoneme recognition in noisy and reverberant speech. In these experiments, the FDLP features derived from high resolution envelopes provide significant improvements.
Full text loading...
Most read this month