DOI: 10.5772/intechopen.

80419

Provisional chapter
Chapter 1

Commonly Used
Some Commonly Used Speech
Speech Feature
FeatureExtraction
Extraction
Algorithms

Sabur Ajibola Alim

Sabur Ajibola Alim and
and Nahrul
Nahrul Khair
Khair Alang
Alang Rashid
Rashid

Additional information isis available

Additional information available at
at the
the end
end of
of the
the chapter
chapter

http://dx.doi.org/10.5772/intechopen.80419

Abstract
Speech is a complex naturally acquired human motor ability. It is characterized in adults
with the production of about 14 different sounds per second via the harmonized actions of
roughly 100 muscles. Speaker recognition is the capability of a software or hardware to
receive speech signal, identify the speaker present in the speech signal and recognize the
speaker afterwards. Feature extraction is accomplished by changing the speech waveform
to a form of parametric representation at a relatively minimized data rate for subsequent
processing and analysis. Therefore, acceptable classification is derived from excellent and
quality features. Mel Frequency Cepstral Coefficients (MFCC), Linear Prediction Coeffi-
cients (LPC), Linear Prediction Cepstral Coefficients (LPCC), Line Spectral Frequencies
(LSF), Discrete Wavelet Transform (DWT) and Perceptual Linear Prediction (PLP) are the
speech feature extraction techniques that were discussed in these chapter. These methods
have been tested in a wide variety of applications, giving them high level of reliability and
acceptability. Researchers have made several modifications to the above discussed tech-
niques to make them less susceptible to noise, more robust and consume less time. In
conclusion, none of the methods is superior to the other, the area of application would
determine which method to select.

Keywords: human speech, speech features, mel frequency cepstral coefficients (MFCC),
linear prediction coefficients (LPC), linear prediction cepstral coefficients (LPCC), line
spectral frequencies (LSF), discrete wavelet transform (DWT), perceptual linear prediction
(PLP)

1. Introduction

Human beings express their feelings, opinions, views and notions orally through speech. The
speech production process includes articulation, voice, and fluency [1, 2]. It is a complex

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons
© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative
Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and eproduction in any medium, provided the original work is properly cited.
distribution, and reproduction in any medium, provided the original work is properly cited.
4 From Natural to Artificial Intelligence - Algorithms and Applications

naturally acquired human motor abilities, a task categorized in regular adults by the production
of about 14 different sounds per second via the harmonized actions of roughly 100 muscles
connected by spinal and cranial nerves. The simplicity with which human beings speak is in
contrast to the complexity of the task, and that complexity could assist in explaining why speech
can be very sensitive to diseases associated with the nervous system [3].
There have been several successful attempts in the development of systems that can analyze,
classify and recognize speech signals. Both hardware and software that have been developed
for such tasks have been applied in various fields such as health care, government sectors and
agriculture. Speaker recognition is the capability of a software or hardware to receive speech
signal, identify the speaker present in the speech signal and recognize the speaker afterwards [4].
Speaker recognition executes a task similar to what the human brain undertakes. This starts
from speech which is an input to the speaker recognition system. Generally, speaker recogni-
tion process takes place in three main steps which are acoustic processing, feature extraction
and classification/recognition [5].
The speech signal has to be processed to remove noise before the extraction of the important
attributes in the speech [6] and identification. The purpose of feature extraction is to illustrate a
speech signal by a predetermined number of components of the signal. This is because all the
information in the acoustic signal is too cumbersome to deal with, and some of the information
is irrelevant in the identification task [7, 8].
Feature extraction is accomplished by changing the speech waveform to a form of parametric
representation at a relatively lesser data rate for subsequent processing and analysis. This is
usually called the front end signal-processing [9, 10]. It transforms the processed speech signal
to a concise but logical representation that is more discriminative and reliable than the actual
signal. With front end being the initial element in the sequence, the quality of the subsequent
features (pattern matching and speaker modeling) is significantly affected by the quality of the
front end [10].

Therefore, acceptable classification is derived from excellent and quality features. In present
automatic speaker recognition (ASR) systems, the procedure for feature extraction has nor-
mally been to discover a representation that is comparatively reliable for several conditions of
the same speech signal, even with alterations in the environmental conditions or speaker,
while retaining the portion that characterizes the information in the speech signal [7, 8].
Feature extraction approaches usually yield a multidimensional feature vector for every
speech signal [11]. A wide range of options are available to parametrically represent the speech
signal for the recognition process, such as perceptual linear prediction (PLP), linear prediction
coding (LPC) and mel-frequency cepstrum coefficients (MFCC). MFCC is the best known and
very popular [9, 12]. Feature extraction is the most relevant portion of speaker recognition.
Features of speech have a vital part in the segregation of a speaker from others [13]. Feature
extraction reduces the magnitude of the speech signal devoid of causing any damage to the
power of speech signal [14].
Before the features are extracted, there are sequences of preprocessing phases that are first
carried out. The preprocessing step is pre-emphasis. This is achieved by passing the signal
 Some Commonly Used Speech Feature Extraction Algorithms 5
http://dx.doi.org/10.5772/intechopen.80419

through a FIR filter [15] which is usually a first-order finite impulse response (FIR) filter [16].
This is succeeded by frame blocking, a method of partitioning the speech signal into frames. It
removes the acoustic interface existing in the start and end of the speech signal [17].
The framed speech signal is then windowed. Bandpass filter is a suitable window [15] that is
applied to minimize disjointedness at the start and finish of each frame. The two most famous
categories of windows are Hamming and Rectangular windows [18]. It increases the sharpness
of harmonics, eliminates the discontinuous of signal by tapering beginning and ending of the
frame zero. It also reduces the spectral distortion formed by the overlap [17].

2. Mel frequency cepstral coefficients (MFCC)

Mel frequency cepstral coefficients (MFCC) was originally suggested for identifying monosyl-
labic words in continuously spoken sentences but not for speaker identification. MFCC com-
putation is a replication of the human hearing system intending to artificially implement the
ear’s working principle with the assumption that the human ear is a reliable speaker recog-
nizer [19]. MFCC features are rooted in the recognized discrepancy of the human ear’s critical
bandwidths with frequency filters spaced linearly at low frequencies and logarithmically at
high frequencies have been used to retain the phonetically vital properties of the speech signal.
Speech signals commonly contain tones of varying frequencies, each tone with an actual
frequency, f (Hz) and the subjective pitch is computed on the Mel scale. The mel-frequency
scale has linear frequency spacing below 1000 Hz and logarithmic spacing above 1000 Hz.
Pitch of 1 kHz tone and 40 dB above the perceptual audible threshold is defined as 1000 mels,
and used as reference point [20].
MFCC is based on signal disintegration with the help of a filter bank. The MFCC gives a
discrete cosine transform (DCT) of a real logarithm of the short-term energy displayed on the
Mel frequency scale [21]. MFCC is used to identify airline reservation, numbers spoken into a
telephone and voice recognition system for security purpose. Some modifications have been
proposed to the basic MFCC algorithm for better robustness, such as by lifting the log-mel-
amplitudes to an appropriate power (around 2 or 3) before applying the DCT and reducing the
impact of the low-energy parts [4].

2.1. Algorithm description, strength and weaknesses

MFCC are cepstral coefficients derived on a twisted frequency scale centerd on human audi-
tory perception. In the computation of MFCC, the first thing is windowing the speech signal to
split the speech signal into frames. Since the high frequency formants process reduced ampli-
tude compared to the low frequency formants, high frequencies are emphasized to obtain
similar amplitude for all the formants. After windowing, Fast Fourier Transform (FFT) is
applied to find the power spectrum of each frame. Subsequently, the filter bank processing is
carried out on the power spectrum, using mel-scale. The DCT is applied to the speech signal
6 From Natural to Artificial Intelligence - Algorithms and Applications

after translating the power spectrum to log domain in order to calculate MFCC coefficients [5].
The formula used to calculate the mels for any frequency is [19, 22]:

melðf Þ ¼ 2595 x log 10 ð1 þ f =700Þ (1)

where mel(f) is the frequency (mels) and f is the frequency (Hz).

The MFCCs are calculated using this equation [9, 19]:

k 
X     
^n ¼ 1 π
C log S^ k cos n k  (2)
n¼1
2 k

where k is the number of mel cepstrum coefficients, S^ k is the output of filterbank and C
^ n is the
final mfcc coefficients.
The block diagram of the MFCC processor can be seen in Figure 1. It summarizes all the
processes and steps taken to obtain the needed coefficients. MFCC can effectively denote the
low frequency region better than the high frequency region, henceforth, it can compute for-
mants that are in the low frequency range and describe the vocal tract resonances. It has been
generally recognized as a front-end procedure for typical Speaker Identification applications,
as it has reduced vulnerability to noise disturbance, with minute session inconsistency and
easy to mine [19]. Also, it is a perfect representation for sounds when the source characteristics
are stable and consistent (music and speech) [23]. Furthermore, it can capture information
from sampled signals with frequencies at a maximum of 5 kHz, which encapsulates most
energy of sounds that are generated by humans [9].
Cepstral coefficients are said to be accurate in certain pattern recognition problems relating to
human voice. They are used extensively in speaker identification and speech recognition [21].
Other formants can also be above 1 kHz and are not efficiently taken into consideration by the
large filter spacing in the high frequency range [19]. MFCC features are not exactly accurate in
the existence of background noise [14, 24] and might not be well suited for generalization [23].

Figure 1. Block diagram of MFCC processor.

 Some Commonly Used Speech Feature Extraction Algorithms 7
http://dx.doi.org/10.5772/intechopen.80419

3. Linear prediction coefficients (LPC)

Linear prediction coefficients (LPC) imitates the human vocal tract [16] and gives robust
speech feature. It evaluates the speech signal by approximating the formants, getting rid of its
effects from the speech signal and estimate the concentration and frequency of the left behind
residue. The result states each sample of the signal as a direct incorporation of previous
samples. The coefficients of the difference equation characterize the formants, thus, LPC needs
to approximate these coefficients [25]. LPC is a powerful speech analysis method and it has
gained fame as a formant estimation method [17].

The frequencies where the resonant crests happen are called the formant frequencies. Thus,
with this technique, the positions of the formants in a speech signal are predictable by calcu-
lating the linear predictive coefficients above a sliding window and finding the crests in the
spectrum of the subsequent linear prediction filter [17]. LPC is helpful in the encoding of high
quality speech at low bit rate [13, 26, 27].
Other features that can be deduced from LPC are linear predication cepstral coefficients
(LPCC), log area ratio (LAR), reflection coefficients (RC), line spectral frequencies (LSF) and
Arcus Sine Coefficients (ARCSIN) [13]. LPC is generally used for speech reconstruction. LPC
method is generally applied in musical and electrical firms for creating mobile robots, in
telephone firms, tonal analysis of violins and other string musical gadgets [4].

3.1. Algorithm description, strength and weaknesses

Linear prediction method is applied to obtain the filter coefficients equivalent to the vocal tract
by reducing the mean square error in between the input speech and estimated speech [28].
Linear prediction analysis of speech signal forecasts any given speech sample at a specific
period as a linear weighted aggregation of preceding samples. The linear predictive model of
speech creation is given as [13, 25]:

X
p
^s ðnÞ ¼ ak sðn  kÞ (3)
k¼1

where ^s is the predicted sample, s is the speech sample, p is the predictor coefficients.
The prediction error is given as [16, 25]:

eðnÞ ¼ sðnÞ  ^s ðnÞ (4)

Subsequently, each frame of the windowed signal is autocorrelated, while the highest autocor-
relation value is the order of the linear prediction analysis. This is followed by the LPC
analysis, where each frame of the autocorrelations is converted into LPC parameters set which
consists of the LPC coefficients [26]. A summary of the procedure for obtaining the LPC is as
seen in Figure 2. LPC can be derived by [7]:
8 From Natural to Artificial Intelligence - Algorithms and Applications

Figure 2. Block diagram of LPC processor.

 
1  km
am ¼ log (5)
1 þ km

where am is the linear prediction coefficient, km is the reflection coefficient.

Linear predictive analysis efficiently selects the vocal tract information from a given speech
[16]. It is known for the speed of computation and accuracy [18]. LPC excellently represents the
source behaviors that are steady and consistent [23]. Furthermore, it is also be used in speaker
recognition system where the main purpose is to extract the vocal tract properties [25]. It gives
very accurate estimates of speech parameters and is comparatively efficient for computation
[14, 26]. Traditional linear prediction suffers from aliased autocorrelation coefficients [29]. LPC
estimates have high sensitivity to quantization noise [30] and might not be well suited for
generalization [23].

4. Linear prediction cepstral coefficients (LPCC)

Linear prediction cepstral coefficients (LPCC) are cepstral coefficients derived from LPC cal-
culated spectral envelope [11]. LPCC are the coefficients of the Fourier transform illustration of
the logarithmic magnitude spectrum [30, 31] of LPC. Cepstral analysis is commonly applied in
the field of speech processing because of its ability to perfectly symbolize speech waveforms
and characteristics with a limited size of features [31].
It was observed by Rosenberg and Sambur that adjacent predictor coefficients are highly
correlated and therefore, representations with less correlated features would be more efficient,
LPCC is a typical example of such. The relationship between LPC and LPCC was originally
derived by Atal in 1974. In theory, it is relatively easy to convert LPC to LPCC, in the case of
minimum phase signals [32].

4.1. Algorithm description, strength and weaknesses

In speech processing, LPCC analogous to LPC, are computed from sample points of a speech
waveform, the horizontal axis is the time axis, while the vertical axis is the amplitude axis [31].
 Some Commonly Used Speech Feature Extraction Algorithms 9
http://dx.doi.org/10.5772/intechopen.80419

Figure 3. Block diagram of LPCC processor.

The LPCC processor is as seen in Figure 3. It pictorially explains the process of obtaining
LPCC. LPCC can be calculated using [7, 15, 33]:

m1  
X k
Cm ¼ a m þ ck amk (6)
k¼1
m

where am is the linear prediction coefficient, Cm is the cepstral coefficient.

LPCC have low vulnerability to noise [30]. LPCC features yield lower error rate as compared
to LPC features [31]. Cepstral coefficients of higher order are mathematically limited, resulting
in an extremely extensive array of variances when moving from the cepstral coefficients of
lower order to cepstral coefficients of higher order [34]. Similarly, LPCC estimates are notori-
ous for having great sensitivity to quantization noise [35]. Cepstral analysis on high-pitch
speech signal gives small source-filter separability in the quefrency domain [29]. Cepstral
coefficients of lower order are sensitive to the spectral slope, while the cepstral coefficients of
higher order are sensitive to noise [15].

5. Line spectral frequencies (LSF)

Individual lines of the Line Spectral Pairs (LSP) are known as line spectral frequencies (LSF).
LSF defines the two resonance situations taking place in the inter-connected tube model of the
human vocal tract. The model takes into consideration the nasal cavity and the mouth shape,
which gives the basis for the fundamental physiological importance of the linear prediction
illustration. The two resonance situations define the vocal tract as either being completely open
or completely closed at the glottis [36]. The two situations begets two groups of resonant
frequencies, with the number of resonances in each group being deduced from the quantity of
linked tubes. The resonances of each situation are the odd and even line spectra correspond-
ingly, and are interwoven into a singularly rising group of LSF [36].
The LSF representation was proposed by Itakura [37, 38] as a substitute to the linear prediction
parametric illustration. In the area of speech coding, it has been realized that this illustration
has an improved quantization features than the other linear prediction parametric illustrations
10 From Natural to Artificial Intelligence - Algorithms and Applications

(LAR and RC). The LSF illustration has the capacity to reduce the bit-rate by 25–30% for
transmitting the linear prediction information without distorting the quality of synthesized
speech [38–40]. Apart from quantization, LSF illustration of the predictor are also suitable for
interpolation. Theoretically, this can be inspired by the point that the sensitivity matrix linking
the LSF-domain squared quantization error to the perceptually relevant log spectrum is diag-
onal [41, 42].

5.1. Algorithm description, strength and weaknesses

LP is established on the point that a speech signal can be defined by Eq. (3). Recall

X
p
^s ðnÞ ¼ ak sðn  kÞ
k¼1

where k is the time index and p is the order of the linear prediction, ^s ðnÞ is the predictor signal
and ak is the LPC coefficients.
The ak coefficients are determined in order to reduce the prediction error by method of
autocorrelation or covariance. Eq. (3) can be modified in the frequency domain with the z-
transform. As such, a small part of the speech signal is anticipated to be given as an output to
the all-pole filter HðzÞ. The new equation is

1 1
H ðzÞ ¼ ¼ P (7)
AðzÞ 1  pi¼1 ai z1

where HðzÞ is the all-pole filter and AðzÞ is the LPC analysis filter.
In order to compute the LSF coefficients, an inverse polynomial filter is split into two poly-
nomials PðzÞ and QðzÞ [36, 38, 40, 41]:

PðzÞ ¼ AðzÞ þ zðpþ1Þ A z1 (8)

QðzÞ ¼ AðzÞ  zðpþ1Þ A z1 (9)

where PðzÞ is the vocal tract with the glottis closed, QðzÞ is the LPC analysis filter of order p.

In order to convert LSF back to LPC, the equation below is used [36, 41, 43, 44]:

AðzÞ ¼ 0:5½PðzÞ þ QðzÞ (10)

The block diagram of the LSF processor is as seen in Figure 4. The most prominent application
of LSF is in the area of speech compression, with extension into the speaker recognition and
speech recognition. This technique has also found restricted use in other fields. LSF have been
investigated for use in musical instrument recognition and coding. LSF have also been applied
to animal noise identification, recognizing individual instruments and financial market analy-
sis. The advantages of LSF include their ability to localize spectral sensitivities, the fact that
 Some Commonly Used Speech Feature Extraction Algorithms 11
http://dx.doi.org/10.5772/intechopen.80419

Figure 4. Block diagram of LSF processor.

they characterize bandwidths and resonance locations and lays emphasis on the important
aspect of spectral peak location. In most instances, the LSF representation provides a near-
minimal data set for subsequent classification [36].
Since LSF represents spectral shape information at a lower data rate than raw input samples, it
is reasonable that a careful use of processing and analysis methods in the LSP domain could
lead to a complexity reduction against alternative techniques operating on the raw input data
itself. LSF play an important role in the transmission of vocal tract information from speech
coder to decoder with their widespread use being a result of their excellent quantization
properties. The generation of LSP parameters can be accomplished using several methods,
ranging in complexity. The major problem revolves around finding the roots of the P and Q
polynomials defined in Eqs. (8) and (9). This can be obtained through standard root solving
methods, or more obscure methods and it is often performed in the cosine domain [36].

6. Discrete wavelet transform (dwt)

Wavelet Transform (WT) theory is centered around signal analysis using varying scales in the
time and frequency domains [45]. With the support of theoretical physicist Alex Grossmann,
Jean Morlet introduced wavelet transform which permits high-frequency events identification
with an enhanced temporal resolution [45–47]. A wavelet is a waveform of effectively limited
duration that has an average value of zero. Many wavelets also display orthogonality, an ideal
feature of compact signal representation [46]. WT is a signal processing technique that can be
used to represent real-life non-stationary signals with high efficiency [33, 46]. It has the ability
to mine information from the transient signals concurrently in both time and frequency
domains [33, 45, 48].

Continuous wavelet transform (CWT) is used to split a continuous-time function into wave-
lets. However, there is redundancy of information and huge computational efforts is
required to calculate all likely scales and translations of CWT, thereby restricting its
use [45]. Discrete wavelet transform (DWT) is an extension of the WT that enhances the
flexibility to the decomposition process [48]. It was introduced as a highly flexible and
efficient method for sub band breakdown of signals [46, 49]. In earlier applications, linear
12 From Natural to Artificial Intelligence - Algorithms and Applications

discretization was used for discretizing CWT. Daubechies and others have developed an
orthogonal DWT specially designed for analyzing a finite set of observations over the set of
scales (dyadic discretization) [47].

6.1. Algorithm description, strength and weaknesses

Wavelet transform decomposes a signal into a group of basic functions called wavelets. Wave-
lets are obtained from a single prototype wavelet called mother wavelet by dilations and
shifting. The main characteristic of the WT is that it uses a variable window to scan the
frequency spectrum, increasing the temporal resolution of the analysis [45, 46, 50].
WT decomposes signals over translated and dilated mother wavelets. Mother wavelet is a time
function with finite energy and fast decay. The different versions of the single wavelet are
orthogonal to each other. The continuous wavelet transform (CWT) is given by [33, 45, 50]:
ð∞  
1 tb
W x ða; bÞ ¼ pffiffiffi xðtÞψ∗ dt (11)
a a
∞

where ψðtÞ is the mother wavelet, a and b are continuous parameters.

The WT coefficient is an expansion and a particular shift represents how well the original signal
corresponds to the translated and dilated mother wavelet. Thus, the coefficient group of CWT
(a, b) associated with a particular signal is the wavelet representation of the original signal in
relation to the mother wavelet [45]. Since CWT contains high redundancy, analyzing the signal
using a small number of scales with varying number of translations at each scale, i.e. discretizing
scale and translation parameters as a ¼ 2j and b ¼ 2j k gives DWT. DWT theory requires two sets
of related functions called scaling function and wavelet function given by [33]:

X
N1 pffiffiffi
ϕðtÞ ¼ h½n 2ϕð2t  nÞ (12)
n¼0

X
N 1 pffiffiffi
ψ ð tÞ ¼ g½n 2ϕð2t  nÞ (13)
n¼0

where ϕðtÞ is the scaling function, ψðtÞ is the wavelet function, h[n] is the an impulse response
of a low-pass filter, and g[n] is an impulse response of a high-pass filter.
There are several ways to discretize a CWT. The DWT of the continuous signal can also be
given by [45]:

ð
þ∞

ðDWT Þðm; pÞ ¼ xðtÞ∙ψm, p dt (14)

∞

where ψm, p is the wavelet function bases, m is the dilation parameter and p is the translation
parameter.
 Some Commonly Used Speech Feature Extraction Algorithms 13
http://dx.doi.org/10.5772/intechopen.80419

Thus, ψm, p is defined as:

 
1 t  pb0 am
ψm, p ¼ pffiffiffiffiffi ψ 0
(15)
am0
am
0

The DWT of a discrete signal is derived from CWT and defined as:
 
1 X n  nb0 am
ðDWT Þðm; kÞ ¼ pffiffiffiffiffi x½n ∙g 0
(16)
am0 n
am
0

where g(*) is the mother wavelet and x[n] is the discretized signal. The mother wavelet may be
dilated and translated discretely by selecting the scaling parameter a ¼ am 0 and translation
parameter b ¼ nb0 am0 (with constants taken as a 0 > 1, b 0 > 1, while m and n are assigned a set
of positive integers).

The scaling and wavelet functions can be implemented effectively using a pair of filters, h[n]
and g[n], called quadrature mirror filters that confirm with the property g½n ¼ ð1Þ1n h½n.
The input signal is filtered by a low-pass filter and high-pass filter to obtain the approximate
components and the detail components respectively. This is summarized in Figure 5. The
approximate signal at each stage is further decomposed using the same low-pass and high-
pass filters to get the approximate and detail components for the next stage. This type of
decomposition is called dyadic decomposition [33].

The DWT parameters contain the information of different frequency scales. This enhances
the speech information obtained in the corresponding frequency band [33]. The ability of the
DWT to partition the variance of the elements of the input on a scale by scale basis is an
added advantage. This partitioning leads to the opinion of the scale-dependent wavelet
variance, which in many ways is equivalent to the more familiar frequency-dependent
Fourier power spectrum [47]. Classic discrete decomposition schemes, which are dyadic do
not fulfill all the requirements for direct use in parameterization. DWT does provide ade-
quate number of frequency bands for effective speech analysis [51]. Since the input signals
are of finite length, the wavelet coefficients will have unwantedly large variations at the
boundaries because of the discontinuities at the boundaries [50].

Figure 5. Block diagram of DWT.

14 From Natural to Artificial Intelligence - Algorithms and Applications

7. Perceptual linear prediction (PLP)

Perceptual linear prediction (PLP) technique combines the critical bands, intensity-to-loudness
compression and equal loudness pre-emphasis in the extraction of relevant information from
speech. It is rooted in the nonlinear bark scale and was initially intended for use in speech
recognition tasks by eliminating the speaker dependent features [11]. PLP gives a representa-
tion conforming to a smoothed short-term spectrum that has been equalized and compressed
similar to the human hearing making it similar to the MFCC. In the PLP approach, several
prominent features of hearing are replicated and the consequent auditory like spectrum of
speech is approximated by an autoregressive all–pole model [52]. PLP gives minimized reso-
lution at high frequencies that signifies auditory filter bank based approach, yet gives the
orthogonal outputs that are similar to the cepstral analysis. It uses linear predictions for
spectral smoothing, hence, the name is perceptual linear prediction [28]. PLP is a combination
of both spectral analysis and linear prediction analysis.

7.1. Algorithm description, strength and weaknesses

In order to compute the PLP features the speech is windowed (Hamming window), the Fast
Fourier Transform (FFT) and the square of the magnitude are computed. This gives the power
spectral estimates. A trapezoidal filter is then applied at 1-bark interval to integrate the
overlapping critical band filter responses in the power spectrum. This effectively compresses
the higher frequencies into a narrow band. The symmetric frequency domain convolution on
the bark warped frequency scale then permits low frequencies to mask the high frequencies,
concurrently smoothing the spectrum. The spectrum is subsequently pre-emphasized to approx-
imate the uneven sensitivity of human hearing at a variety of frequencies. The spectral ampli-
tude is compressed, this reduces the amplitude variation of the spectral resonances. An Inverse
Discrete Fourier Transform (IDCT) is performed to get the autocorrelation coefficients. Spec-
tral smoothing is performed, solving the autoregressive equations. The autoregressive coeffi-
cients are converted to cepstral variables [28]. The equation for computing the bark scale
frequency is:

Figure 6. Block diagram of PLP processor.

 Type of Shape of What is Speed of Type of Noise Sensitivity to Reliability Frequency
Filter filter modeled computation coefficient resistance quantization/additional captured
noise

Mel frequency cepstral Mel Triangular Human High Cepstral Medium Medium High Low
coefficient (MFCC) Auditory
System
Linear prediction Linear Linear Human Vocal High Autocorrelation High High High Low
coefficient (LPC) Prediction Tract Coefficient
Linear prediction cepstral Linear Linear Human Vocal Medium Cepstral High High Medium Low &
coefficient (LPCC) Prediction Tract Medium
Line spectral frequencies Linear Linear Human Vocal Medium Spectral High High Medium Low &
(LSF) Prediction Tract Medium
Discrete wavelet Lowpass & — — High Wavelets Medium Medium Medium Low &
transform (DWT) highpass High
Perceptual linear Bark Trapezoidal Human Medium Cepstral & Medium Medium Medium Low &
prediction (PLP) Auditory Autocorrelation Medium
System

Table 1. Comparison between the feature extraction techniques.

http://dx.doi.org/10.5772/intechopen.80419
Some Commonly Used Speech Feature Extraction Algorithms
15
16 From Natural to Artificial Intelligence - Algorithms and Applications

26:81 f
barkðf Þ ¼  0:53 (17)
1960 þ f

where bark(f) is the frequency (bark) and f is the frequency (Hz).

The identification achieved by PLP is better than that of LPC [28], because it is an improvement
over the conventional LPC because it effectively suppresses the speaker-dependent informa-
tion [52]. Also, it has enhanced speaker independent recognition performance and is robust to
noise, variations in the channel and microphones [53]. PLP reconstructs the autoregressive
noise component accurately [54]. PLP based front end is sensitive to any change in the formant
frequency.

Figure 6 shows the PLP processor, showing all the steps to be taken to obtain the PLP
coefficients. PLP has low sensitivity to spectral tilt, consistent with the findings that it is
relatively insensitive to phonetic judgments of the spectral tilt. Also, PLP analysis is dependent
on the result of the overall spectral balance (formant amplitudes). The formant amplitudes are
easily affected by factors such as the recording equipment, communication channel and addi-
tive noise [52]. Furthermore, the time-frequency resolution and efficient sampling of the short-
term representation are addressed in an ad-hoc way [54].
Table 1 shows a comparison between the six feature extraction techniques that have been
explicitly described above. Even though the selection of a feature extraction algorithm for use
in research is individual dependent, however, this table has been able to characterize these
techniques based on the main considerations in the selection of any feature extraction algo-
rithm. The considerations include speed of computation, noise resistance and sensitivity to
additional noise. The table also serves as a guide when considering the selection between any
two or more of the discussed algorithms.

8. Conclusion

MFCC, LPC, LPCC, LSF, PLP and DWT are some of the feature extraction techniques used for
extracting relevant information form speech signals for the purpose speech recognition and
identification. These techniques have stood the test of time and have been widely used in
speech recognition systems for several purposes. Speech signal is a slow time varying signal,
quasi-stationary, when observed over an adequately short period of time between 5 and
100 msec, its behavior is relatively stationary. As a result of this, short time spectral analysis
which includes MFCC, LPCC and PLP are commonly used for the extraction of important
information from speech signals. Noise is a serious challenge encountered in the process of
feature extraction, as well as speaker recognition as a whole. Subsequently, researchers have
made several modifications to the above discussed techniques to make them less susceptible to
noise, more robust and consume less time. These methods have also been used in the recogni-
tion of sounds. The extracted information will be the input to the classifier for identification
purposes. The above discussed feature extraction approaches can be implemented using
MATLAB.
 Some Commonly Used Speech Feature Extraction Algorithms 17
http://dx.doi.org/10.5772/intechopen.80419

Author details

Sabur Ajibola Alim1* and Nahrul Khair Alang Rashid2

*Address all correspondence to: [email protected]

1 Ahmadu Bello University, Zaria, Nigeria

2 Universiti Teknologi Malaysia, Skudai, Johor, Malaysia

References

[1] Hariharan M, Vijean V, Fook CY, Yaacob S. Speech stuttering assessment using sample
entropy and Least Square Support vector machine. In: 8th International Colloquium on
Signal Processing and its Applications (CSPA). 2012. pp. 240-245

[2] Manjula GN, Kumar MS. Stuttered speech recognition for robotic control. International
Journal of Engineering and Innovative Technology (IJEIT). 2014;3(12):174-177

[3] Duffy JR. Motor speech disorders: Clues to neurologic diagnosis. In: Parkinson’s Disease
and Movement Disorders. Totowa, NJ: Humana Press; 2000. pp. 35-53

[4] Kurzekar PK, Deshmukh RR, Waghmare VB, Shrishrimal PP. A comparative study of
feature extraction techniques for speech recognition system. International Journal of Inno-
vative Research in Science, Engineering and Technology. 2014;3(12):18006-18016

[5] Ahmad AM, Ismail S, Samaon DF. Recurrent neural network with backpropagation
through time for speech recognition. In: IEEE International Symposium on Communica-
tions and Information Technology (ISCIT 2004). Vol. 1. Sapporo, Japan: IEEE; 2004. pp. 98-
102

[6] Shaneh M, Taheri A. Voice command recognition system based on MFCC and VQ algo-
rithms. World academy of science. Engineering and Technology. 2009;57:534-538

[7] Mosa GS, Ali AA. Arabic phoneme recognition using hierarchical neural fuzzy petri net
and LPC feature extraction. Signal Processing: An International Journal (SPIJ). 2009;3(5):
161

[8] Yousefian N, Analoui M. Using radial basis probabilistic neural network for speech
recognition. In: Proceeding of 3rd International Conference on Information and Knowl-
edge (IKT07), Mashhad, Iran. 2007

[9] Cornaz C, Hunkeler U, Velisavljevic V. An Automatic Speaker Recognition System. Swit-

zerland: Lausanne; 2003. Retrieved from: http://read.pudn.com/downloads60/sourcecode/
multimedia/audio/209082/asr_project.pdf

[10] Shah SAA, ul Asar A, Shaukat SF. Neural network solution for secure interactive voice
response. World Applied Sciences Journal. 2009;6(9):1264-1269
18 From Natural to Artificial Intelligence - Algorithms and Applications

[11] Ravikumar KM, Rajagopal R, Nagaraj HC. An approach for objective assessment of
stuttered speech using MFCC features. ICGST International Journal on Digital Signal
Processing, DSP. 2009;9(1):19-24

[12] Kumar PP, Vardhan KSN, Krishna KSR. Performance evaluation of MLP for speech
recognition in noisy environments using MFCC & wavelets. International Journal of
Computer Science & Communication (IJCSC). 2010;1(2):41-45

[13] Kumar R, Ranjan R, Singh SK, Kala R, Shukla A, Tiwari R. Multilingual speaker recogni-
tion using neural network. In: Proceedings of the Frontiers of Research on Speech and
Music, FRSM. 2009. pp. 1-8

[14] Narang S, Gupta MD. Speech feature extraction techniques: A review. International Jour-
nal of Computer Science and Mobile Computing. 2015;4(3):107-114

[15] Al-Alaoui MA, Al-Kanj L, Azar J, Yaacoub E. Speech recognition using artificial neural
networks and hidden Markov models. IEEE Multidisciplinary Engineering Education
Magazine. 2008;3(3):77-86

[16] Al-Sarayreh KT, Al-Qutaish RE, Al-Kasasbeh BM. Using the sound recognition techniques to
reduce the electricity consumption in highways. Journal of American Science. 2009;5(2):1-12

[17] Gill AS. A review on feature extraction techniques for speech processing. International
Journal Of Engineering and Computer Science. 2016;5(10):18551-18556

[18] Othman AM, Riadh MH. Speech recognition using scaly neural networks. World academy
of science. Engineering and Technology. 2008;38:253-258

[19] Chakroborty S, Roy A, Saha G. Fusion of a complementary feature set with MFCC for
improved closed set text-independent speaker identification. In: IEEE International Con-
ference on Industrial Technology, 2006. ICIT 2006. pp. 387-390

[20] de Lara JRC. A method of automatic speaker recognition using cepstral features and
vectorial quantization. In: Iberoamerican Congress on Pattern Recognition. Berlin, Heidel-
berg: Springer; 2005. pp. 146-153

[21] Ravikumar KM, Reddy BA, Rajagopal R, Nagaraj HC. Automatic detection of syllable
repetition in read speech for objective assessment of stuttered Disfluencies. In: Proceed-
ings of World Academy Science, Engineering and Technology. 2008. pp. 270-273

[22] Hasan MR, Jamil M, Rabbani G, Rahman MGRMS. Speaker Identification Using Mel
Frequency cepstral coefficients. In: 3rd International Conference on Electrical & Computer
Engineering, 2004. ICECE 2004. pp. 28-30

[23] Chu S, Narayanan S, Kuo CC. Environmental sound recognition using MP-based features.
In: IEEE International Conference on Acoustics, Speech and Signal Processing, 2008.
ICASSP 2008. IEEE; 2008. pp. 1-4

[24] Rao TB, Reddy PPVGD, Prasad A. Recognition and a panoramic view of Raaga emotions
of singers-application Gaussian mixture model. International Journal of Research and
Reviews in Computer Science (IJRRCS). 2011;2(1):201-204
 Some Commonly Used Speech Feature Extraction Algorithms 19
http://dx.doi.org/10.5772/intechopen.80419

[25] Agrawal S, Shruti AK, Krishna CR. Prosodic feature based text dependent speaker recog-
nition using machine learning algorithms. International Journal of Engineering Science
and Technology. 2010;2(10):5150-5157

[26] Paulraj MP, Sazali Y, Nazri A, Kumar S. A speech recognition system for Malaysian
English pronunciation using neural network. In: Proceedings of the International Confer-
ence on Man-Machine Systems (ICoMMS). 2009

[27] Tan CL, Jantan A. Digit recognition using neural networks. Malaysian Journal of Com-
puter Science. 2004;17(2):40-54

[28] Kumar P, Chandra M. Speaker identification using Gaussian mixture models. MIT Inter-
national Journal of Electronics and Communication Engineering. 2011;1(1):27-30

[29] Wang TT, Quatieri TF. High-pitch formant estimation by exploiting temporal change of
pitch. IEEE Transactions on Audio, Speech, and Language Processing. 2010;18(1):171-186

[30] El Choubassi MM, El Khoury HE, Alagha CEJ, Skaf JA, Al-Alaoui MA. Arabic speech
recognition using recurrent neural networks. In: Proceedings of the 3rd IEEE International
Symposium on Signal Processing and Information Technology (IEEE Cat. No.03EX795).
Ieee; 2003. pp. 543-547. DOI: 10.1109/ISSPIT.2003.1341178

[31] Wu QZ, Jou IC, Lee SY. On-line signature verification using LPC cepstrum and neural
networks. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics. 1997;
27(1):148-153

[32] Holambe R, Deshpande M. Advances in Non-Linear Modeling for Speech Processing.

Berlin, Heidelberg: Springer Science & Business Media; 2012

[33] Nehe NS, Holambe RS. DWT and LPC based feature extraction methods for isolated word
recognition. EURASIP Journal on Audio, Speech, and Music Processing. 2012;2012(1):7

[34] Young S, Evermann G, Gales M, Hain T, Kershaw D, Liu X, et al. The HTK Book, Version 3.4.
Cambridge, United Kingdom: Cambridge University; 2006

[35] Ismail S, Ahmad A. Recurrent neural network with backpropagation through time algo-
rithm for arabic recognition. In: Proceedings of the 18th European Simulation Multiconference
(ESM). Magdeburg, Germany; 2004. pp. 13-16

[36] McLoughlin IV. Line spectral pairs. Signal Processing. 2008;88(3):448-467

[37] Itakura F. Line spectrum representation of linear predictor coefficients of speech signals.
The Journal of the Acoustical Society of America. 1975;57(S1):S35-S35

[38] Silva DF, de Souza VM, Batista GE, Giusti R. Spoken digit recognition in portuguese using
line spectral frequencies. Ibero-American Conference on Artificial Intelligence. Vol. 7637.
Berlin, Heidelberg: Springer; 2012. pp. 241-250

[39] Kabal P, Ramachandran RP. The computation of line spectral frequencies using
Chebyshev polynomials. IEEE Transactions on Acoustics, Speech and Signal Processing.
1986;34(6):1419-1426
20 From Natural to Artificial Intelligence - Algorithms and Applications

[40] Paliwal KK. On the use of line spectral frequency parameters for speech recognition.
Digital Signal Processing. 1992;2(2):80-87
[41] Alang Rashid NK, Alim SA, Hashim NNWNH, Sediono W. Receiver operating character-
istics measure for the recognition of stuttering Dysfluencies using line spectral frequen-
cies. IIUM Engineering Journal. 2017;18(1):193-200
[42] Kleijn WB, Bäckström T, Alku P. On line spectral frequencies. IEEE Signal Processing
Letters. 2003;10(3):75-77
[43] Bäckström T, Pedersen CF, Fischer J, Pietrzyk G. Finding line spectral frequencies using
the fast Fourier transform. In: 2015 IEEE International Conference on in Acoustics, Speech
and Signal Processing (ICASSP). 2015. pp. 5122-5126
[44] Nematollahi MA, Vorakulpipat C, Gamboa Rosales H. Semifragile speech watermarking
based on least significant bit replacement of line spectral frequencies. Mathematical Prob-
lems in Engineering. 2017. 9 p
[45] Oliveira MO, Bretas AS. Application of discrete wavelet transform for differential protec-
tion of power transformers. In: IEEE PowerTech. Bucharest: IEEE; 2009. pp. 1-8

[46] Gupta D, Choubey S. Discrete wavelet transform for image processing. International
Journal of Emerging Technology and Advanced Engineering. 2015;4(3):598-602

[47] Lindsay RW, Percival DB, Rothrock DA. The discrete wavelet transform and the scale
analysis of the surface properties of sea ice. IEEE Transactions on Geoscience and Remote
Sensing. 1996;34(3):771-787

[48] Turner C, Joseph A. A wavelet packet and mel-frequency cepstral coefficients-based

feature extraction method for speaker identification. In: Procedia Computer Science.
2015. pp. 416-421
[49] Reig-Bolaño R, Marti-Puig P, Solé-Casals J, Zaiats V, Parisi V. Coding of biosignals using
the discrete wavelet decomposition. In: International Conference on Nonlinear Speech
Processing. Berlin Heidelberg: Springer; 2009. pp. 144-151
[50] Tufekci Z, Gowdy JN. Feature extraction using discrete wavelet transform for speech
recognition. In: IEEE Southeastcon 2000. 2000. pp. 116-123
[51] Gałka J, Ziółko M. Wavelet speech feature extraction using mean best basis algorithm. In:
International Conference on Nonlinear Speech Processing Berlin. Heidelberg: Springer;
2009. pp. 128-135
[52] Hermansky H. Perceptual linear predictive (PLP) analysis of speech. The Journal of the
Acoustical Society of America. 1990;87(4):1738-1752
[53] Picone J. Fundamentals of Speech Recognition: Spectral Transformations. 2011. Retrieved
from: http://www.isip.piconepress.com/publications/courses/msstate/ece_8463/lectures/
current/lecture_17/lecture_17.pdf
[54] Thomas S, Ganapathy S, Hermansky H. Spectro-temporal features for automatic speech
recognition using linear prediction in spectral domain. In: Proceedings of the 16th European
Signal Processing Conference (EUSIPCO 2008), Lausanne, Switzerland. 2008