A Baseline System for Speaker Recognition

1
A Baseline System for Speaker Recognition

• C. Mokbel, H. Greige, R. Zantout, H. Abi Akl
• A. Ghaoui, J. Chalhoub, R. Bayeh
• University Of Balamand - ELISA

2
Outline

• Introduction
• Baseline speaker recognition system
• NIST 2002 evaluation
• Conclusion and perspective

3
Introduction

• A baseline system has been built and was used in
  the NIST 2002 speaker recognition evaluation
• GMM based system
• Normalization using z-norm
• Adaptation technique used to estimate speaker
  model starting from world model

4
Baseline Speaker Recognition System

• Feature extraction
• Speech recognition based feature vectors
• 13 MFCC coefficients including the energy on
  logarithmic scale
• first and second order derivative
• Leading to 39 feature parameters
• Preprocessing using cepstral mean normalization

5
Baseline Speaker Recognition System

• GMM modeling for both hypotheses speaker and non
  speaker (world)
• EM algorithm to train the world model
  (Baum-Welch)
• Initialization using LBG VQ
• Speaker model adapted mean vectors from the
  world model
• Approximation of the unified adaptation
  approach (Online Adaptation of HMMs to
  Real-Life Conditions A Unified Framework, IEEE
  Trans. on SAP Vol. 9, n 4, may 2001) IEEE Trans.
  on SAP Vol. 9, n 4, may 2001)

6
Baseline Speaker Recognition System

• Speaker Adaptation
• World model Gaussian distributions grouped in a
  binary tree
• Speaker data driven determination of the Gaussian
  classes
• MLLR applied based on these classes only means
  of Gaussian distributions are adapted
• MAP applied to the leaves Gaussian distributions

7
Baseline Speaker Recognition System

• Building the Gaussian tree bottom up
• Grouping two by two the closest Gaussian
  distributions
• Distance between 2 Gaussian distributions is
  equal to the loss in the likelihood of the
  associated data if the two Gaussian are merged in
  a unique Gaussian

8
Baseline Speaker Recognition System

• After the E-step of the EM algorithm the weights
  associated to the leaves of the tree are
  propagated through the tree up to the root
• Going from the root to the leaves, nodes are
  selected whenever one of their two children has a
  weight less than a threshold
• This defines a partition that will be used in an
  MLLR algorithm

9
Baseline Speaker Recognition System

• MAP algorithm
• Estimated Gaussian means parameters at the leaves
  are smoothed using a fixed weight with the
  parameters of the world Gaussian

10
Baseline Speaker Recognition System

• Given a target speaker model ls, the world model
  lw and a test utterance X, the score for this
  utterance is computed as the log likelihood
  ratio
• s log p(X/ls) / p(X/lw)
• This score should be normalized due to the fact
  that the world model is not precise

11
Baseline Speaker Recognition System

• Normalization using the z-norm
• Few impostors utterances are used
• A score is computed for every utterance
• The different scores define a distribution per
  target speaker
• Target speakers distributions should be similar
  for a decision using a unique threshold
• Reduce and center the distribution
• ns a s b

12
Baseline Speaker Recognition System

• Based on the data from the 2001 evaluation a DET
  curve can be plotted
• Find the optimal decision threshold that minimize
  the cost defined by NIST2002, i.e.
• Cdet CmisPrmiss/targetPrtarget
  CFalseAlarmPrFalseAlarm/NonTarget(1-Prtarget)

13
NIST 2002 evaluation

• Feature vector 13 MFCCs 13 d 13 d2
• Cepstral Mean Normalization
• Gender dependent GMM with 256 Gaussian mixtures
  for world model
• Trained on a subset of the cellular data of NIST
  2001 evaluation

14
NIST 2002 evaluation

• Target speaker model adapted from world model
• For every iteration and after the E step
• Threshold (cumulative probability 3.0) to
  select tree nodes
• MLLR used to update the Gaussian means
• Approximated MAP to smooth the MLLR estimated
  parameters linear combination between the MLLR
  estimated mean (0.8) and the world (a priori)
  mean (0.2)

15
NIST 2002 evaluation

• 16 male and 21 female speakers (NIST 2001) used
  as impostors (8 test files from each)
• The pseudo-impostors scores define a distribution
  used to z-normalize the score for a given target
  speaker
• Global threshold estimated on NIST 2001 data in
  order to minimize the cost

16
NIST 2002 evaluation

• System characteristics
• CPU time on a pentium III 800 MHz
• 2.1 ms per frame and per speaker for speaker
  model adaptation
• 0.92 ms per frame for the test
• Memory usage
• 360 Kbytes per test

17
NIST 2002 evaluation

• Results
• Cdet 0.100292
• Min Cdet 0.097833
• DET Curve

18
NIST 2002 evaluation
19
NIST 2002 evaluation
20
NIST 2002 evaluation
21
Conclusions and perspectives

• A new baseline system has been developed and
  evaluated
• A lot of work to be done, mainly
• Optimize the feature extraction module
• Implement the complete Unified Adaptation
  approach
• Investigate new normalization strategies
• Integrate automatic labeling of speech segments