Missing Feature Imputation of Log-spectral Data for Noise Robust Asr
نویسندگان
چکیده
In this paper, we present a missing feature (MF) imputation algorithm for log-spectral data with applications to noise robust ASR. Drawing from previous work [1], we adapt the previously proposed spectrographic reconstruction solution to the liftered log-spectral domain by introducing log-spectral flooring (LS-FLR). LS-FLR is shown to be an efficient and effective noise robust feature extraction technique. When LSFLR is integrated in deriving the novel log-spectral data imputation framework, the overall system is shown to provide significant improvements in noise robust speech recognition.
منابع مشابه
Mask estimation in non-stationary noise environments for missing feature based robust speech recognition
In missing feature based automatic speech recognition (ASR), the role of the spectro-temporal mask in providing an accurate description of the relationship between target speech and environmental noise is critical for minimizing the degradation in ASR word accuracy (WAC) as the signal-to-noise ratio (SNR) decreases. This paper demonstrates the importance of accurate characterization of instanta...
متن کاملSome solution to the missing feature problem in data classification, with application to noise robust ASR
We address the theoretical and practical issues involved in ASR when some of the observation data for the target signal is masked by other signals. Techniques discussed range from simple missing data imputation to Bayesian optimal classification. We have developed the Bayesian approach because this allows prior knowledge to be incorporated naturally into the recognition process, thereby permitt...
متن کاملSpectral Reconstruction and Noise Model Estimation Based on a Masking Model for Noise Robust Speech Recognition
An effective way to increase noise robustness in automatic speech recognition (ASR) systems is feature enhancement based on an analytical distortion model that describes the effects of noise on the speech features. One of such distortionmodels that has been reported to achieve a good trade-off between accuracy and simplicity is the masking model. Under this model, speech distortion caused by en...
متن کاملMMSE Feature Reconstruction Based on an Occlusion Model for Robust ASR
This paper proposes a novel compensation technique developed in the log-spectral domain. Our proposal consists in a minimum mean square error (MMSE) estimator derived from an occlusion model [1]. According to this model, the effect of noise over speech is simplified to a binary masking, so that the noise is completely masked by the speech when the speech power dominates and the other way round ...
متن کاملLow Complexity Spectral Imputation for Noise Robust Speech Recognition
of the Thesis Low Complexity Spectral Imputation for Noise Robust Speech Recognition by Julien van Hout Master of Science in Electrical Engineering University of California, Los Angeles, 2012 Professor Abeer Alwan, Chair With the recent push of Automatic Speech Recognition (ASR) capabilities to mobile devices, the user’s voice is now recorded in environments with a potentially high level of bac...
متن کامل