Using sparse representations for missing data imputation in noise robust speech recognition
Jort F. Gemmeke,Bert Cranen +1 more
- 01 Jan 2008
- pp 1-5
TL;DR: A novel imputation technique working on entire words that achieves recognition accuracies of 92% at SNR -5 dB using oracle masks on AURORA-2 as compared to 61% using a conventional frame-based approach.
read more
Abstract: Noise robustness of automatic speech recognition benefits from using missing data imputation: Prior to recognition the parts of the spectrogram dominated by noise are replaced by clean speech estimates. Especially at low SNRs each frame contains at best only a few uncorrupted coefficients. This makes frame-by-frame restoration of corrupted feature vectors error-prone, and recognition accuracy will mostly be sub-optimal. In this paper we present a novel imputation technique working on entire words. A word is sparsely represented in an overcomplete basis of exemplar (clean) speech signals using only the uncorrupted time-frequency elements of the word. The corrupted elements are replaced by estimates obtained by projecting the sparse representation in the basis. We achieve recognition accuracies of 92% at SNR −5 dB using oracle masks on AURORA-2 as compared to 61% using a conventional frame-based approach. The performance obtained with estimated masks can be directly related to the proportion of correctly identified uncorrupted coefficients.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Sparse Representation for Computer Vision and Pattern Recognition
John Wright,Yi Ma,Julien Mairal,Guillermo Sapiro,Thomas S. Huang,Shuicheng Yan +5 more
- 29 Apr 2010
TL;DR: This review paper highlights a few representative examples of how the interaction between sparse signal representation and computer vision can enrich both fields, and raises a number of open questions for further study.
Block-Sparse Recovery via Convex Optimization
Ehsan Elhamifar,René Vidal +1 more
TL;DR: It is shown that treating the face recognition problem as a block-sparse recovery problem improves the state-of-the-art results by 10% with only 25% of the training data.
193
SAR Target Recognition via Joint Sparse Representation of Monogenic Signal
TL;DR: The development of three component-specific feature descriptors for each monogenic component is produced first and the resulting features are fed into a joint sparse representation model to exploit the intercorrelation among multiple tasks.
185
Sparse Representation for Computer Vision and Pattern Recognition
John Wright,Yue Ma,Julien Mairal,Guillermo Sapiro,Thomas S. Huang,Shuicheng Yan +5 more
- 01 May 2009
Compressive Sensing for Missing Data Imputation in Noise Robust Speech Recognition
TL;DR: This paper introduces a novel non-parametric, exemplar-based method for reconstructing clean speech from noisy observations, based on techniques from the field of Compressive Sensing, which can impute missing features using larger time windows such as entire words.
References
Regression Shrinkage and Selection via the Lasso
TL;DR: A new method for estimation in linear models called the lasso, which minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant, is proposed.
Least angle regression
Bradley Efron,Trevor Hastie,Iain M. Johnstone,Robert Tibshirani,Hemant Ishwaran,Keith Knight,Jean-Michel Loubes,Jean-Michel Loubes,Pascal Massart,Pascal Massart,David Madigan,David Madigan,Greg Ridgeway,Greg Ridgeway,Saharon Rosset,Saharon Rosset,Ji Zhu,Robert A. Stine,Berwin A. Turlach,Sanford Weisberg +19 more
TL;DR: A publicly available algorithm that requires only the same order of magnitude of computational effort as ordinary least squares applied to the full set of covariates is described.
Least Angle Regression
TL;DR: Least Angle Regression (LARS) as discussed by the authors is a new model selection algorithm, which is a useful and less greedy version of traditional forward selection methods such as All Subsets, Forward Selection and Backward Elimination.
Sparse Approximate Solutions to Linear Systems
TL;DR: It is shown that the problem is NP-hard, but that the well-known greedy heuristic is good in that it computes a solution with at most at most $\left\lceil 18 \mbox{ Opt} ({\bf \epsilon}/2) \|{\bf A}^+\|^2_2 \ln(\|b\|_2/{\bf
3.2K
For most large underdetermined systems of linear equations the minimal 1-norm solution is also the sparsest solution
TL;DR: In this article, the authors consider linear equations y = Φx where y is a given vector in ℝn and Φ is a n × m matrix with n 0 so that for large n and for all Φ's except a negligible fraction, the solution x1of the 1-minimization problem is unique and equal to x0.
3K
Related Papers (5)
Bradley Efron,Trevor Hastie,Iain M. Johnstone,Robert Tibshirani,Hemant Ishwaran,Keith Knight,Jean-Michel Loubes,Jean-Michel Loubes,Pascal Massart,Pascal Massart,David Madigan,David Madigan,Greg Ridgeway,Greg Ridgeway,Saharon Rosset,Saharon Rosset,Ji Zhu,Robert A. Stine,Berwin A. Turlach,Sanford Weisberg +19 more