Dataset shift compensation in machine learning

Patent metadata

Field	Value
Publication number	US-2016019883-A1
Application number	US-201414331230-A
Country	US
Kind code	A1
Filing date	Jul 15, 2014
Priority date	Jul 15, 2014
Publication date	Jan 21, 2016
Grant date	—

Abstract

A method for inter-dataset variability compensation, the method comprising using at least one hardware processor for: receiving a heterogeneous development dataset comprising multiple samples and metadata associated with at least some of the multiple samples; dividing the multiple samples into multiple homogenous subsets, based on the metadata; averaging high-level features of each of the multiple homogenous subsets, to produce multiple central high-level features for the multiple homogenous subsets, respectively; computing an inter-dataset variability subspace spanned by the multiple central high-level features; removing the inter-dataset variability subspace from the high-level features of the multiple homogenous subsets, to produce denoised samples; and training a machine learning system using the denoised speech samples.

First claim

What is claimed is:

1 . A method for inter-dataset variability compensation, the method comprising using at least one hardware processor for:
receiving a heterogeneous development dataset comprising multiple samples and metadata associated with at least some of the multiple samples; dividing the multiple samples into multiple homogenous subsets, based on the metadata; computing a statistical measure of high-level features of each of the multiple homogenous subsets, to produce multiple central high-level features for the multiple homogenous subsets, respectively; computing an inter-dataset variability subspace spanned by the multiple central high-level features; removing the inter-dataset variability subspace from the high-level features of the multiple homogenous subsets, to produce denoised samples; and training a machine learning system using the denoised speech samples.

2 . The method according to claim 1 , wherein the high-level features are selected from the group consisting of:
i-vectors, GMM (Gaussian Mixture Model) supervectors, HMM (Hidden Markov Model) supervectors, d-vectors, JFA (Joint Factor Analysis) supervectors, LBP (Local Binary Patterns), HOG (Histograms of Oriented Gradients), and EBIF (Early Biologically-Inspired Features).

3 . The method according to claim 2 , wherein the machine learning system is selected from the group consisting of:
a PLDA (Probabilistic Linear Discriminant Analysis)-based system, an SVM (Support Vector Machine)-based system, a neural network-based system, a NAP (Nuisance Attribute Projection)-based system, a WCCN (Within-Speaker Covariance Matrix)-based system, and an LDA (Linear Discriminant Analysis)-based system.

4 . The method according to claim 3 , wherein the multiple samples are speech samples.

5 . The method according to claim 4 , wherein the heterogeneous development dataset is devoid of speech samples from a target domain of the speaker recognition.

6 . The method according to claim 4 , wherein the metadata comprises at least one parameter selected from the group consisting of: speaker gender, spoken language and recordation setting.

7 . The method according to claim 4 , wherein the computing of the inter-dataset variability subspace comprises PCA (Principal Component Analysis).

8 . A method for inter-dataset variability compensation for speaker recognition, the method comprising using at least one hardware processor for:
receiving a heterogeneous development dataset comprising multiple speech samples; dividing the multiple speech samples into multiple homogenous subsets; for each subset i of the multiple homogenous subsets: (a) estimating PLDA (Probabilistic Linear Discriminant Analysis) hyper-parameters {μ i , B i , W i }, wherein p denotes a center of an i-vector space, B denotes a between-speaker covariance matrix and W denotes a within-speaker covariance matrix, and (b) computing an i-vector subspace S μ corresponding to {μ i }, an i-vector subspace S W corresponding to {W i }, and an i-vector subspace S B corresponding to {B i }; joining i-vector subspaces S μ , S W and S B into a single subspace S; removing subspace S from i-vectors of the multiple speech samples, to produce denoised speech samples; and training a PLDA speaker recognition system using the denoised speech samples.

9 . The method according to claim 8 , further comprising smoothing B by linear interpolation using an estimated diagonal of B.

10 . The method according to claim 8 , wherein:
the heterogeneous development dataset further comprises metadata associated with at least some of the multiple speech samples; and the dividing is based on the metadata.

11 . The method according to claim 8 , wherein the computing of each of the i-vector subspaces S μ , S W and S B comprises PCA (Principal Component Analysis).

12 . The method according to claim 8 , further comprising computing an average of squared {W i }, and finding a k number of largest eigenvalues of the squared {W i }, wherein the k largest eigenvalues span the i-vector subspace S W .

13 . The method according to claim 12 , further comprising whitening the i-vector subspace S W with respect to W.

14 . The method according to claim 8 , further comprising computing an average of squared {B i }, and finding an m number of largest eigenvalues of the squared {B i }, wherein the k largest eigenvalues span the i-vector subspace S B .

15 . The method according to claim 14 , further comprising whitening the i-vector subspace S B with respect to B.

16 . A computer program product for inter-dataset variability compensation for speaker recognition, the computer program product comprising a non-transitory computer-readable storage medium having program code embodied therewith, the program code executable by at least one hardware processor to:
receive a heterogeneous development dataset comprising multiple speech samples; divide the multiple speech samples into multiple homogenous subsets; for each subset i of the multiple homogenous subsets: (a) estimate PLDA (Probabilistic Linear Discriminant Analysis) hyper-parameters {μ i , B i , W i }, wherein μ denotes a center of an i-vector space, B denotes a between-speaker covariance matrix and W denotes a within-speaker covariance matrix, and (b) compute an i-vector subspace S μ corresponding to {μ i }, an i-vector subspace S W corresponding to {W i }, and an i-vector subspace S B corresponding to {B i }; join i-vector subspaces S μ , S W and S B into a single subspace S; remove subspace S from i-vectors of the multiple speech samples, to produce denoised speech samples; and train a PLDA speaker recognition system using the denoised speech samples.

17 . The computer program product according to claim 16 , wherein the program code is further executable by the at least one hardware processor to smooth B by linear interpolation using an estimated diagonal of B.

18 . The computer program product according to claim 16 , wherein:
the heterogeneous development dataset further comprises metadata associated with at least some of the multiple speech samples; and the dividing is based on the metadata.

19 . The computer program product according to claim 16 , wherein the computing of each of the i-vector subspaces S μ , S W and S B comprises PCA (Principal Component Analysis).

20 . The computer program product according to claim 16 , wherein the program code is further executable by the at least one hardware processor to:
compute an average of squared {W,}; find a k number of largest eigenvalues of the squared {W i }, wherein the k largest eigenvalues span the i-vector subspace S W ; whiten the i-vector subspace S W with respect to W; compute an average of squared {B i }; find an m number of largest eigenvalues of the squared {B i }, wherein the m largest eigenvalues span the i-vector subspace S b ; and whiten the i-vector subspace S B with respect to B.

Assignees

• IBM

Inventors

• Aronowitz Hagai

Classifications

• G10L17/26

  Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices · CPC title

• G10L21/0208

  Noise filtering · CPC title

• G10L17/04

  Training, enrolment or model building · CPC title

• G10L15/20

  Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech (G10L21/02 takes precedence) · CPC title

• G10L17/20

  Pattern transformations or operations aimed at increasing system robustness, e.g. against channel noise or different working conditions · CPC title