RCCMean Subtraction Robust Feature and Compare Various Feature based Methods for Robust Speech Recog

1
RCC-Mean Subtraction Robust Feature and Compare
Various Feature based Methods for Robust Speech
Recognition in presence of Telephone Noise

• Amin Fazel
• Sharif University of Technology
• Hossein Sameti, Mohammad T. Manzuri
• February 2005

Computer Engineering Department, Sharif
University of Technology
2
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

3
Effect of Noise on ASR

• Two phase in most ASR systems
• Train
• Operating (Testing)
• Mismatch causes reduction in accuracy
• Mismatch occur because of
• Environment
• Microphone, babble, distance, transmission canal
• Speaker
• Specific speaker speed,
• Various speakers gender, age, accent,

4
Effect of Noise on ASR

• Noise
• Additive noise
• Babble, car, subway
• Exhibit, office,
• Convolutional Noise
• Canal, telephone line
• Microphone effect
• Distance of speaker to microphone
• Others
• Lombard noise, Reflection of building

5
Effect of Noise on ASR

• Simple model
• Robust Speech Recognition is the study of
  building speech recognition that handle mismatch
  condition.

6
Robustness Methods

• Signal
• Speech enhancement
• Feature
• Robust feature extraction
• Model
• Change of the model parameters
• Model training

7
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

8
Mel-Frequency Cepstral Coefficient

• Compute magnitude-squared of Fourier transform
• Apply triangular frequency weights that represent
  the effects of peripheral auditory frequency
  resolution
• Take log of outputs ( for RCC we take root
  instead of log)
• Compute cepstral using discrete cosine transform
• Smooth by dropping higher-order coefficients

9
Temporal processing

• To capture the temporal features of the spectral
  envelop to provide the robustness
• Delta Feature first and second order
  differences regression
• Cepstral Mean Subtraction
• For normalizing for channel effects and adjusting
  for spectral slope

10
Perceptual Linear Prediction (PLP)

• Compute magnitude-squared of Fourier transform
• Apply triangular frequency weights that represent
  the effects of peripheral auditory frequency
  resolution
• Apply compressive nonlinearities
• Compute discrete cosine transform
• Smooth using autoregressive modeling
• Compute cepstral using linear recursion

11
PLP (Cont.)

• Algorithm

12
RelAtive SpecTral Analysis

• Which makes PLP (and possibly also some other
  short-term spectrum based techniques) more robust
  to linear spectral distortions
• The new spectral estimate is less sensitive to
  slow variations in the short-term spectrum
• Filtering of the temporal trajectories of some
  function of each of the spectral values to
  provide more reliable spectral features
• This is usually a bandpass filter, maintaining
  the linguistically important spectral envelop
  modulation (1-16Hz)

13
RASTA (Cont.)

• Algorithm

14
RASTA-PLP

• Algorithm

15
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

16
RCC-Mean Normalization

• Root Cepstral Coefficients (RCC)
• Derived using root compression rather than log
  compression on the filterbank energies
• Advantage of RCC to MFCC
• More immune to noise
• Faster decoding

17
RCC-Mean Normalization

• Mean normalization
• If we approximate root with logarithm

18
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

19
Experiment 1

• Database
• TFARSDAT
• 64 Speakers
• 8 hours telephony speech data
• ASR
• Sharif ASR System
• HMM based
• Training Segmental K-means
• Search Beam Viterbi

20
Experiment 1

• Test results

21
Experiment 2

• Aurora 2.0
• Noisy connected digits recognition
• 4 hours training data, 2 hours test data in 70
  Noise Types/SNR conditions
• HTK
• HMM based
• Model for each digit
• 16 states with 3 Gaussian mixtures

22
Experiment 2

• Average results on AURORA
• Average obtained on various SNRs of a noise

23
Experiment 2

• Subway noise in various SNRs

24
Experiment 2

• Babble noise in various SNRs

25
Experiment 2

• Car noise in various SNRs

26
Experiment 2

• Exhibition noise in various SNRs

27
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

28
Summery

• Various robust features was tested
• Introduce of RCC_MN
• In first experiment
• RASTA-PLP
• Although RCC_MN is good
• In second experiment
• RCC_MN

29
Outline

• Introduction
• Feature based methods
• MFCC, RCC, CMN, PLP, RASTA
• Mean Normalization Root Cepstral Coefficients
• Experimental Results
• Experiment 1 Sharif CSR and TFARSDAT Database
• Experiment 2 HTK CSR and AURORA 2 Database
• Summery

30
Thanks for your patience !