The implementation of channel diversity in mobile software radio receivers

Schnepp, H.; Müller, Thomas; Luy, J.-F.; Russer, P.

doi:10.1109/lmwc.2003.815704

Cited by 1 publication

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Low-band and highband signal representations are inferred from the transmitted region, assuming a high degree of correlation, and are added to the input signal to reconstruct a wideband signal [1,2]. Similar studies exist for in-vehicle communications and other noisy environments [3,4]. Some of these studies have considered performing experiments using reconstructed speech with an extended frequency bandwidth; however, there has been limited literature specifically addressing the issue of bandwidth extension to improve ASR rates.…”

Section: Introductionmentioning

confidence: 99%

MFCC Compensation for Improved Recognition of Filtered and Band-Limited Speech

Morales

Hansen

Toledano³

Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005.

View full text Add to dashboard Cite

This paper addresses the problem of bandwidth expansion for the purpose of robust speech recognition. We show that an HMM-based ASR engine trained with full spectrum range data (0-8kHz) can successfully perform speech recognition tasks over band-filtered test data compensated by means of a series of simple MFCC parameter corrector functions. The problem is important when ASR is employed for audio streams of unknown frequency bandwidth common in spoken document retrieval. Evaluation is based on recognition rates. Accuracy varies depending on the width and spectral regions eliminated, but the system shows great advantages over the use of uncompensated filtered test data. The theoretical maximum recognition rates using corrector functions over filtered test data are very close to the base rate (unfiltered data) even when the greatest part of the spectrum of the original data is suppressed. These rates are even better than those obtained in the matched train/test HMMs with filtered data.

show abstract

Section: Introductionmentioning

confidence: 99%