HIWIRE MEETING Torino, March 910, 2006

1
HIWIRE MEETINGTorino, March 9-10, 2006

• José C. Segura, Javier Ramírez

2
Schedule

• HIWIRE database evaluations
• New results HEQ and PEQ
• Non-linear feature normalization
• Using temporal redundancy
• HEQ integration in Loquendo platform
• Recursive estimation of the equalization function
• New improvements in robust VAD
• Bispectrum-based VAD
• SVM-enabled VAD

3
HIWIRE database evaluations
4
Schedule

• HIWIRE database evaluations
• New results HEQ and PEQ
• Non-linear feature normalization
• Using temporal redundancy
• HEQ integration in Loquendo platform
• Recursive estimation of the equalization function
• New improvements in robust VAD
• Bispectrum-based VAD
• SVM-enabled VAD

5
Temporal redundancy in HEQ

• Enhance the normalization adding a linear
  transformation to restore temporal correlations
• Each equalized cepstral coefficient is
  time-filtered with an ARMA filter that restores
  the autocorrelation of clean data

6
HEQ integration in Loquendo platform
7
HEQ integration (recursive estimation) (1)

• Actual approach Gaussian HEQ using ECDF

• Using quantiles

8
HEQ integration (recursive estimation) (2)

• Equalization by linear interpolation

Averaged over training data
From actual utterance

• Mapping correspondingquantiles

9
HEQ integration (recursive estimation) (3)
10
HEQ integration (recursive estimation) (4)

• Utterances are equalized WITHOUT delay
• Quantiles are updated AFTER the equalization

11
HIWIRE MEETINGTorino, March 9-10, 2006

• José C. Segura, Javier Ramírez

12
Schedule

• HIWIRE database evaluations
• New results HEQ and PEQ
• Non-linear feature normalization
• Using temporal redundancy
• HEQ integration in Loquendo platform
• Recursive estimation of the equalization function
• New improvements in robust VAD
• Bispectrum-based VAD
• SVM-enabled VAD

13
Bispectrum-based VAD (1)

• Motivations
• Ability of HOS methods to detect signals in noise
  
• Knowledge of the input processes (Gaussian)
• Issues to be addressed
• Computationally expensive
• Variance of bispectrum estimators much higher
  than that of power spectral estimators (identical
  data record size)
• Solution Integrated bispectrum
• J. K. Tugnait, Detection of non-Gaussian signals
  using integrated polyspectrum, IEEE Trans. on
  Signal Processing, vol. 42, no. 11, pp.
  31373149, 1994.

14
Bispectrum-based VAD (2)

• Definitions
• Let x(t) be a discrete-time signal
• Bispectrum
• Third order cumulants
• Inverse transform

15
Bispectrum-based VAD (3)
Noise only
Noise speech
16
Bispectrum-based VAD (4)

• Integrated bispectrum (IBI)
• Cross-spectrum Syx(?)
• Bispectrum
• Inverse
• transform
• Bispectrum Cross spectrum

i 0
17
Bispectrum-based VAD (5)

• Integrated bispectrum (IBI)
• Defined as a cross spectrum between the signal
  and its square, and therefore, it is a function
  of a single frequency variable
• Benefits
• Less computational cost
• computed as a cross spectrum
• Variance of the same order of the power spectrum
  estimator
• Properties
• For Gaussian processes
• Bispectrum is zero
• Integrated bispectrum is zero as well

18
Bispectrum-based VAD (6)

• Two alternatives explored for formulating the
  decision rule
• Estimation by block averaging (BA)
• MO-LRT
• Given a set of N 2m1 consecutive observations

19
Bispectrum-based VAD (7)

• LRT evaluation
• IBI Gaussian Model

• Variances
• Defined in terms of
• Sss (clean speech power spectrum)
• Snn (noise power spectrum)

20
Bispectrum-based VAD (8)

• Denoising

2nd WF stage
1st WF stage
2nd WF design
Smoothed spectral subtraction
1st WF design
1-frame delay
21
Bispectrum VAD Analysis (1)

• MO-LRT VAD

22
Bispectrum-based VAD results (2)
23
Bispectrum-based VAD results (3)
24
Bispectrum-based VAD results (4)
WF Wiener filtering FD Frame-dropping
25
SVM-enabled VAD (1)

• Motivation
• Ability of SVMs for learning from experimental
  data
• SVMs enable defining a function
• using training data
• Classify unseen examples (x, y)
• Statistical learning theory restricts the class
  of functions the learning machine can implement.

26
SVM-enabled VAD (2)

• Hyperplane classifiers
• Training w and b define maximal margin
  hyperplane
• Kernels

27
SVM-enabled VAD (3)
28
SVM-enabled VAD (4)

• Feature
• extraction
• Training

29
SVM-enabled VAD (5)

• Feature
• extraction
• Decision function
• 2-band features

30
SVM-enabled VAD (6)

• Analysis
• 4 subbands
• Noise reduction
• Improvements
• Contextual speech features
• Better results without noise reduction

31
Dissemination (VAD)

• Integrated bispectrum
• J.M. Górriz, J. Ramírez, C. G. Puntonet, J.C.
  Segura, Generalized-LRT based voice activity
  detector, IEEE Signal Processing Letters, 2006.
• J. Ramírez , J.M. Górriz, J. C. Segura, C. G.
  Puntonet, A. Rubio, Speech/Non-speech
  Discrimination based on Contextual Information
  Integrated Bispectrum LRT, IEEE Signal
  Processing Letters, 2006.
• J.M. Górriz, J. Ramírez, J. C. Segura, C. G.
  Puntonet, L. García, Effective Speech/Pause
  Discrimination Using an Integrated Bispectrum
  Likelihood Ratio Test , ICASSP 2006.
• SVM VAD
• J. Ramírez, P. Yélamos, J.M. Górriz, J.C. Segura.
  SVM-based Speech Endpoint Detection Using
  Contextual Speech Features, IEE Electronics
  Letters 2006.
• J. Ramírez, P. Yélamos, J.M. Górriz, C.G.
  Puntonet, J.C. Segura. SVM-enabled Voice
  Activity Detection, ISNN 2006.
• P. Yélamos, J. Ramírez, J.M. Górriz, C.G.
  Puntonet, J.C. Segura, Speech Event Detection
  Using Support Vector Machines, ICCS 2006.

32
HIWIRE MEETINGAthens, November 3-4, 2005

• José C. Segura, Javier Ramírez