01 8445 speech enhancement

International Journal of Informatics and Communication Technology (IJ-ICT)
Vol.9, No.2, August 2020, pp. 63~72
ISSN: 2252-8776, DOI: 10.11591/ijict.v9i2.pp63-72  63
Journal homepage: http://ijict.iaescore.com
Speech enhancement using modified wiener filter based MMSE
and speech presence probability estimation
V. Vijayasri Bolisetty1
, U. Yedukondalu2
, I. Santiprabha3
1
Department of Electronics and Communication Engineering, Aditya College of Engineering and Technology, India
2
Electronics & Communication Engineering Department, Sri Vasavi Engineering College, India
3
Department of Electronics & Communication Engineering, University College of Engineering, JNTUK, India
Article Info ABSTRACT
Article history:
Received Mar 6, 2019
Revised Dec 12, 2019
Accepted Jan 12, 2020
In the present-day communications speech signals get contaminated due to
various sorts of noises that degrade the speech quality and adversely impacts
speech recognition performance. To overcome these issues, a novel approach
for speech enhancement using Modified Wiener filtering is developed and
power spectrum computation is applied for degraded signal to obtain the
noise characteristics from a noisy spectrum. In next phase, MMSE technique
is applied where Gaussian distribution of each signal i.e. original and noisy
signal is analyzed. The Gaussian distribution provides spectrum estimation
and spectral coefficient parameters which can be used for probabilistic model
formulation. Moreover, a-priori-SNR computation is also incorporated for
coefficient updation and noise presence estimation which operates similar to
the conventional VAD. However, conventional VAD scheme is based on the
hard threshold which is not capable to derive satisfactory performance and a
soft-decision based threshold is developed for improving the performance of
speech enhancement. An extensive simulation study is carried out using
MATLAB simulation tool on NOIZEUS speech database and a comparative
study is presented where proposed approach is proved better in comparison
with existing technique.
Keywords:
A-priori-SNR
Gaussian distribution
Noise tracking
PSD
SPP
This is an open access article under the CC BY-SA license.
Corresponding Author:
V. Vijayasri Bolisetty,
Department of Electronics and Communication Engineering,
Aditya College of Engineering and Technology,
Aditya Nagar, ADB Rd, Surampalem, Andhra Pradesh, 533437, India.
Email: vasudha.tweety@gmail.com
1. INTRODUCTION
Now-a-days, speech communications have drawn demand with the advent of different broadband
and multimedia applications in rapid usage at different environments. In such scenarios, preserving the
information in speech signal plays an important role which has impact on the efficient communication.
Moreover, due to low-cost digital signal processors and memory chips, speech processing applications, voice
communication systems and voice recognition systems are widely utilized in various real-time application
scenarios [1]. Various communication devices also have been manufactured i.e. mobile phones and mics etc.
which can be used for indoor as well as outdoor applications. In these environments, background noise is
present which gets accumulated with the original signal resulting degraded speech quality and it may lead to
the performance degradation of voice recognition systems [2].
In outdoor environments like heavy traffic, bus terminals, malls, etc., presence of various types of
background noises contaminate the original signal that leads to improper communication [3] in real-time
applications like speaker recognition system, mobile communication, hearing aids etc. Hence noise is to be

 ISSN: 2252-8776
Int J Inf & Commun Technol, Vol. 9, No. 2, August 2020: 63 – 72
64
estimated and reduced to improve the speech quality. During last decade, various enhancement techniques
have been presented for speech signal enhancement. Substantially, these techniques are categorized single-
channel and multi-channel speech enhancement. Usually, back ground noise features can be identified using
single channel speech enhancement and the effects of reverberation can be reduced through multichannel
speech enhancement. Though, Multichannel schemes show significant performance for speech enhancement
when compared to single-channel speech enhancement, the enhancement is to be done through each
microphone individually. Hence the work is concise to single channel speech enhancement.
The speech enhancement algorithm aims to rectify the damaged input or output signal and to
increase the performance of the communication link. The damaged speech signal generates huge trouble
mainly for speech recognition applications. The quality and intelligibility of speech are damaged by the noise
involved in the speech signal. The term intelligibility denotes the understandability of final outcome of the
speech signal. The accuracy of the exact content of speech signal is termed as quality. The different types of
speech enhancement algorithm used to deduct noise are Adaptive or non-adaptive, frequency or time domain
[4]. Over the last years, several techniques have been presented for speech signal enhancement i.e. Spectral
Subtraction, Wavelet based methods, model-based techniques and filtering techniques [5, 6]. Furthermore,
speech enhancement techniques can be categorized as spectral and temporal processing techniques.
According to the spectral domain process, corrupted signal is processed through the transform domain
technique whereas temporal processing method uses time-domain analysis for improving the quality of the
speech signal.
The complete article is organized as follows: section 2 presents a brief literature survey and recent
advancements in speech enhancement techniques. Section 3 presents proposed speech enhancement system
modeling. Results are analyzed and discussion is elaborated in section 4, finally, concluding remarks are
presented in the section 5.
1. LITERATURE SURVEY
Conventional multiband speech enhancement acquires two operations: one is splitting the spectrum
into frequency bands, and the other is executing speech enhancement in each band independently. The polo-
interaction problem in the spectral domain leads to the suppression of few coefficients in the estimation of
clean speech by the influence of the formants in the neighboring bands and thus grades in poor quality. To
reduce the domination of stronger formants over the neighboring bands, the assessment of clean speech is
done by, in the temporal domain. The unsuppressed speech is filtered into various equivalent rectangular
bandwidth based subbands and followed by enhancement of spectral speech in each band based on Discrete
Cosine Transform (DCT) using Spectral Subtraction/Minimum Mean Square Error (MMSE) [7].
Park et. al. [8] discussed about the use of speech enhancement technique in mobile communication
systems. Authors developed an efficient scheme for noise reduction which can improve the performance of
speech recognition. However, mobile devices have limited capacity which motivates to develop a low
complex scheme for noise reduction. In this work, a speech coder is also utilized for packet data estimation.
In general, this work uses pitch information for comb filtering. Adaptive filtering technique has a significant
impact on the speech processing system. Adaptive filtering has been used widely in the various applications
such as channel equalization, system identification and echo cancellation. Although, MMSE and DWT shows
improvement in speech enhancement, their performance is poor at low SNR conditions [9, 10].
The Dual Tree Complex Wavelet Packet Transform (DTCWPT) employed in [11] provides a
solution to avoid the aliasing and oscillations due to shift invariance in DWT. Even though compatibility
exists between MMSE based SPP and DTCWPT, the loss of intelligibility in the reconstructed speech due to
the finite wavelet filters. In this field, Choi. et. al. [12] improved the existing adaptive filtering and developed
a novel sub-band adaptive filter. Further, this technique takes the advantage of norm-optimization and norm
as the computation of cost function. Authors have claimed that this technique is capable to improve the
system performance and robustness in impulsive noise scenarios.
The optimally modified MMSE based log Spectral Amplitude estimator (OM-LSA) improves the
performance compared to the MMSE in case of stationary and some sort of non-stationary noises but shows
significantly poor performance at low and high SNR conditions [13, 14]. SMPO based MMSE technique for
speech enhancement improves the PESQ in all kinds of noisy environments. Binary masking results good
speech quality but sometimes intelligibility of the speech may be lost due to over thresholds in masking noise
coefficients. SMPO-Weiner is going to provide better solution in these issues [15]. Hence combination of
Weiner filter with MMSE based SPP through soft decision may improve the speech intelligibility in addition
to speech quality.
− Challenges in Speech Enhancement: Speech is considered as powerful mode of communication not only
for humans but also for human machine interface. Sometimes the speech signal travels through the noisy

Int J Inf & Commun Technol ISSN: 2252-8776 
Speech enhancement using modified wiener filter based MMSE and speech … (V. Vijayasri Bolisetty)
65
medium before reaching to the listener (recognition system). Now-a-days, automated speech processing
systems have gained lot of attraction from researchers and have been adopted widely in the real-life
scenarios. State-of-art models of automated speech processing systems works well with controlled
environment but real-time systems suffer from various background noises, reverberation and speech
from other speakers, which causes partial loss of information to complete loss especially at low SNR.
Noise removal can be done only by identifying the characteristics of the noise through preprocessing.
However, huge number of researches have been carried out in this field but due to computational
complexities, speech enhancement still remains a challenging task as the process itself introduces
complexity.
− Contribution of the work. Main contributions of this work are as follows:
a. Initially, the features of the speech signal, like silence is estimated from the noisy signal spectrum
through Speech Presence Probability (SPP) estimation.
b. Extracting the noise characteristics from the silence.
c. Finally, MMSE (Noise Tracker) based noise power estimation to suppress the noisy signal using
noise power tracking scheme.
2. SINGLE-CHANNEL SPEECH ENHANCEMENT
Assuming that the majority of the transmission noise is additive, the speech enhancement for single-
channel is represented in (1),
𝑦(𝑛) = 𝑥(𝑛) + 𝑑(𝑛) (1)
where x(n), d(n), y(n) corresponds to speech, noise and noisy speech signals and n indicates discrete time
index. Both x(n) and d(n) are independent with zero mean value. The block diagram of single channel speech
enhancement is represented in Figure 1. To estimate the clean speech signal x(n), the noisy signal y(n) is
processed using speech enhancement algorithms.
Figure 1. General Speech Enhancement Systems
An analysis of speech enhancement method for a noisy speech signal y(n) is illustrated in Figure 2.
Initially, the noisy signal is segmented into overlapping frames, then transformed into frequency domain
using DFT or STFT. To achieve good quality of speech signal, DFT technique is applied as it easily
understands the spectral content of the signal.
Figure 2. Analysis – synthesis of a noisy speech signal

 ISSN: 2252-8776
66
Prior to the speech enhancement technique, the noise characteristics can be identified by the
statistical model. The gain values are calculated based on the features extracted from the original input so that
an upgraded speech signal is acquired. The target evaluation can be made through the power spectrum
density of the noisy speech segments and noise. This helps to generate the gain coefficients of the signals.
Wiener filtering helps to reduce the Mean Square Error (MSE) among the actual signal and the
estimated signal and to boost up the original signal from the noisy signal. Similarly, the noise can be
minimized through hard or soft thresholding. Sometimes the important components of the original speech
might be lost due to over thresholding.
3.1. Modified wiener filtering technique
The advantage of the modified adaptive wiener filter is that the speech signal is processed through
filter by varying local statistics such as mean, variance. Here, the mean value of the additive noise d(n) is
considered to be zero and holds a white nature with variance σd
2
. Thus, the power spectrum Pd(ω) can be
expressed in (2),
𝑃𝑑(𝜔) = 𝜎 𝑑
2
(2)
Here the segmented speech signal xj(n) is treated as stationary and thus can be modeled by (3),
𝑥𝑗(𝑛) = 𝑚 𝑥𝑗 + 𝜎𝑥𝑗 𝑤(𝑛) (3)
where mxj is the local mean and σxj is the standard deviation of xj(n). w(n) is a unit variance noise with a zero
mean. For convenience xj is represented as x, the mean of the original signal ms is equal to average mean
value of all the j frames, mx. The Wiener filter transfer function can be estimated by (4),
𝐻(𝜔) =
𝑃 𝑋(𝜔)
𝑃 𝑋(𝜔)+𝑃 𝑑(𝜔)
=
𝜎 𝑋
2
𝜎 𝑋
2+𝜎 𝑑
2 (4)
The impulse response of the wiener filter can be achieved by applying Inverse Transformation to
H(ω) and is given in (5),
ℎ(𝑛) =
𝜎 𝑋
2
𝜎 𝑋
2+𝜎 𝑑
2 𝛿(𝑛) (5)
As the mx and σx are updated at each sample, the speech signal can be estimated from (6), and is
denoted as 𝑥̂(n) .
𝑥
^
(𝑛) = 𝑚 𝑥(𝑛) +
𝜎 𝑥
2(𝑛)
𝜎 𝑥
2(𝑛)+𝜎 𝑑
2 (𝑥(𝑛) − 𝑚 𝑥(𝑛)) (6)
Now estimating and tracking the noise frames in the noisy signal is the crucial task in the process of
speech enhancement. Usually, the voice activity detector is employed for finding the presence of noise. But
in this Weiner filtering method, frame energy is synchronized with the minimum frame energy. Minimum
energy variation is directly proportional to the signal conditions. Therefore, by using smoothed 32 points of
the spectrum the spectral deviance is calculated. To gain noise spectrum these points are verified and
upgraded additionally. Then the level of energy is fixed to 10dB. So by analyzing the energy level which is
more than 10 dB and RMS level more than 8 dB, the presence of speech can be identified. When there is no
availability of speech this noise spectrum N(ω,m) can be measured for varied time samples. The ultimate plan
is to implement MMSE based noise power estimation technique. In this process, speech & noise spectral
coefficients have complex Gaussian distribution which can be expressed in (7).
𝑃𝐷(𝑛) =
1
𝜎 𝐷
2 𝛱
𝑒𝑥𝑝 (−
|𝑛|2
𝜎 𝑥
2 ) (7)
Hence, the noisy power spectral coefficients can be expressed as given in (8).
𝑃𝑦(𝑥) =
1
𝜎 𝐷
2(1+𝜉)
𝑒𝑥𝑝 (−
|𝑥|2
𝜎 𝐷
2(1+𝜉)
) (8)

67
3.2. Noise PSD estimation and tracking
The noise is represented as 𝑛 = 𝐷𝑒 𝑗𝛥
and 𝑥 = 𝑅𝑒 𝑗𝜃
. Spectral coefficients can be transformed into
polar coordinates using (9).
𝑝 𝐷,𝛥(𝑑, 𝛿) =
1
𝜎 𝐷
2 𝜋
𝑒𝑥𝑝 (−
𝑑2
𝜎 𝐷
2) (9)
Probability distribution 𝑝 𝑥|𝐷,𝛥(𝑥|(𝑑,𝛿)) of signal is given in (10).
𝑝 𝑥|𝐷,𝛥(𝑥|𝑑,𝛿) =
1
𝜋𝜎𝑠
2 𝑒𝑥𝑝 (
2𝑑𝑟 𝑐𝑜𝑠(𝛿−𝜃)−𝑟2−𝑑2
𝜎𝑠
2 ) (10)
In the MMSE method there are several noise power estimators present in the noisy signal
periodogram estimation. Let us consider a priori SNR ξ and estimated noise power is 𝜎 𝑑
2
. The noise
periodogram can be obtained in (11).
|𝑛̂|2
= 𝐸(|𝑛|2
𝑥) = (
1
1+𝜉̂) |𝑥|2
+
𝜉̂
1+𝜉̂ 𝜎̂ 𝑑
2
(11)
From the (11) it is identified that the noise periodogram can be gained. But here the signal gets
altered from time to time, therefore the spectral density has to be revised frequently. To update the spectral
density parameters the recursive smoothing is applied as shown in (12).
𝜎̂ 𝑑
2
(𝑙) = 𝛼𝜎̂ 𝑑
2
(𝑙 − 1) + (1 − 𝛼)|𝐷(𝑙)|2
(12)
Where α = 0.0. Whenever ξ < 1, the noise periodogram is updated. In the same way by verifying a
priori SNR factors the spectral noise power is improved. Hence the sum of the observed noisy signal and
previous estimation of spectral noise power 𝜎̂ 𝑁2 is represented by MMSE. From this estimation the priori
SNR value is obtained which is in between 0 and 1. In A priori SNR can be obtained by following (13).
𝜉̂ = 𝑚𝑎𝑥( 0, 𝜉̂ 𝑚𝑙
) = 𝑚𝑎𝑥( 0, 𝛾̂ − 1) (13)
where ml denotes maximum likelihood computation and𝛾̂(𝑙) =
|𝑥(𝑙)|2
𝜎 𝑛
2(𝑙−1). The (13) is unbias in nature. Hence, a
priori SNR information is not known. So, the estimation of this unknown SNR can generate abias factor as
expressed in (14).
𝛣−1
(𝜉) = ((1 + 𝜉)𝛾 (2,
1
1+𝜉
) + 𝑒
−
1
1+𝜉) (14)
where 𝛾 represents incomplete gamma function. According to the nature of a priori SNR, estimator E(|d|2
x)
value is unbiased whereas for low SNR values it is under-biased.
3.3. Soft–decision technique for noise presence estimation
Speech presence or absence is to be identified, to identify the noise characteristics. Generally, noise
characteristics can be obtained when the speech is absent. The speech or silence part classification can be
done by (15).
|𝑑̂(𝑙)|
2
= 𝐸(|𝑑̂(𝑙)|
2
𝑥(𝑙)) = {
𝜎̂ 𝑑
2
(𝑙 − 1), 𝑖𝑓𝛾̂(𝑙) ≥ 1
|𝛾(𝑙)|2
, 𝑖𝑓𝛾̂(𝑙) < 1
(15)
However, proposed model utilizes soft decision-based framework for noise estimation in the
presence or absence of speech. The probability of speech presence or absence can be estimated from (16)
and (17).
𝑝(𝐻1|𝜁)| 𝑃 𝑍|(𝐻1|𝜁)>>𝑃 𝑍|(𝐻0|𝜁)
= 1 (16)
𝑝(𝐻1|𝜁)| 𝑃 𝑍|(𝐻1|𝜁)<<𝑃 𝑍|(𝐻0|𝜁)
= 0 (17)

 ISSN: 2252-8776
68
where,
𝑝𝑧|𝛨0
(𝜁) =
1
𝛤(𝜐)
𝜐 𝜐
𝜁 𝜐−1
𝑒𝑥𝑝( − 𝜐𝜁) (18)
𝑝𝑧|𝛨1
(𝜁) =
1
𝛤(𝜐)
(
𝜐
1+𝜉𝛨1
) 𝜐
𝜁 𝜐−1
𝑒𝑥𝑝( − 𝜐
𝜁
1+𝜉𝐻1
) (19)
The (18) and (19) describes speech presence and speech absence respectively. Noise periodogram
under speech presence conditions can be expressed in (20). For further improvement chi-square distribution-
based hypothesis analysis can be applied.
𝐸(|𝑛̂(𝑙)|2|𝑥) = 𝑃(𝛨0|𝑌)𝐸(|𝑛̂(𝑙)|2|𝑥, 𝛨0) + 𝑃(𝛨1|𝑥)𝐸(|𝑛|2|𝑥, 𝛨1) (20)
There are two important components to be considered in estimating speech presence probability
using chi-square approximation: observation sequence and estimated noise variance as shown in (22).
Spectral component computation is applied resulting in observation sequence and estimated noise variance
generation where total N number of frequency bins is considered given in (21).
𝜊 = [𝜊1, 𝜊2, . . . , 𝜊 𝑘, . . . , 𝜊 𝑁] (21)
𝜀 = [𝑒1, 𝑒2, . . . , 𝑒 𝑘, . . . , 𝑒 𝑁]
In next phase, chi-square calculation is applied for these frequency bins and the chi-square statistic
is given by (22).
𝑁𝑆2
= ∑
(𝜊 𝑘−𝑒 𝑘)2
𝑒 𝑘
𝑁
𝑘=1 (22)
Through the chi-square calculation the obtained value is compared with the threshold parameter
where (N-1) total degrees of frequency bins are available.
Condition:
If (Ns2
>Th)// threshold value is denoted by Th
I(λ,k)=1 // accept condition H1
Else
I(λ,k)=0 // accept condition H0
End
3. RESULTS AND DISCUSSION
The noise characteristics are estimated from the noisy speech signal during the absence of speech
and the speech clearness is determined from Performance Evaluation of Speech Quality (PESQ) and
segmented SNR (segSNR). In the aspect of mobile application, a sample-rate of 16 KHz, 8KHz TIMIT data
base signals are used as input to measure the quality of the proposed algorithm. Initially, the signals are TF-
decomposed and Hann-windowed frames with a length of 256 samples. But to calculate the objective
intelligibility, each frame is zero-padded up to 512 samples. As the PESQ is highly correlated with the
subjective measures, it is recommended by ITU-T for measuring the performance of the processing
technique. The simulation has been done in MATLAB 2013b tool.
The noisy signals such as babble, airport, car, train noise are chosen at different SNR 0dB, 5 dB, 10
dB with 30 samples of male and female speakers from NOIZEUS database are chosen for performance
evaluation of the proposed technique. By using the overlap-and-add procedure the signal is synthesized.
General simulation parameters like down-sampling, Number of FFT points, window length and window
overlap are described in Table 1.
Table 1. simulation parameters considered
Simulation Parmeter Name Simulation Parameter Vlaue
Signal Downsampling 8 kHz
NFFT 2^14
Window Length(ms) 32 ms
Window overlap 16 ms
Number of averaging window 10
PSD (Power Spectral density) estimation parameter 0.8

69
Figure 3. Original speech signal Figure 4. Noisy Signal
Figure 5. Power spectra Figure 6. Filtered Signal and noise suppressed signal
Table 2. Comparative performance for babble noise
Input SNR
Parameter
Name
OM-LSA
[13]
SMPO
[11]
MMSE-SPP
[10]
BWT
[16]
Wavelet
Packet [11]
Proposed
0 dB
PESQ 2.06 2.01 2.08 2.25 2.11 2.37
segSNR 3.62 4.00 2.22 0.96 5.25 5.51
5 dB
PESQ 2.42 2.38 2.47 2.29 2.44 2.52
segSNR 3.59 3.62 1.50 2.55 4.65 4.73
10 dB
PESQ 2.86 2.83 2.87 2.50 2.87 2.99
segSNR 3.10 4.11 0.76 3.27 4.23 4.7
The original speech, noisy signals are considered for simulation in .wav form and is represented in
Figure 3 and Figure 4 respectively. Through these diagrammatical presentations the presence and absence
can be identified. The probability distribution utilizes the phase analysis method and Figure 5. shows the
Power spectra of original, noisy signals. Finally, filtered and noise suppressed signal analysis is represented
in Figure 6.
The proposed model is compared with the other existing state of art (Pengfei Sun, Jun Qin [11]).
PESQ and segSNR are the characteristics features which are used to evaluate the processing approach. By
varying the noisy input SNR values from 0dB to 10dB the average values of PESQ and segSNR values are
represented in the table for all samples. The Noizeus database with 30 speakers in each set contains babble
noise, airport noise, car noise and train noise which is also treated as important characteristics to identify the
performance of the method. The Table 2, Table 3, Table 4, and Table 5 describe the performance analysis for
babble, airport, car and train noises respectively.
The calculated PESQ and segmental SNR are related with benchmark algorithms to estimate the
efficiency of the system. In terms of babble and train noises of 0dB, 5dB and 10 dB SNR signals of existing
state-of-art method the proposed method achieves enhanced results by means of babble and train noises 0dB,

 ISSN: 2252-8776
70
5dB and 10 dB SNR signals. Also, better results are obtained in presence of airport and car noises for all
signals which are shown in Table 3 and 4. From the results it is clearly identified that noise detection is more
efficient than the other existing approaches. The stationary noise are recognized easily whereas in some case
the non-stationary noise like babble noise and train noise tracked with the help of soft decision based
estimation.
Table 3. Comparative performance for airport noise
Input SNR
Parameter
Name
OM-LSA
[13]
SMPO
[11]
MMSE-SPP
[10]
BWT
[16]
Wavelet
Packet [11]
Proposed
0 dB
PESQ 2.13 1.98 2.13 2.10 2.22 2.35
segSNR 3.34 4.71 4.45 1.07 5.15 5.36
5 dB
PESQ 2.78 2.58 2.72 2.41 2.72 2.7
segSNR 2.10 4.12 3.67 2.71 4.91 4.86
10 dB
PESQ 2.88 2.80 2.79 2.65 2.86 2.98
segSNR 1.10 3.71 2.96 2.82 4.30 4.36
Table 4. Comparative performance for car noise
Input SNR
Parameter
Name
OM-LSA
[13]
SMPO [11]
MMSE-SPP
[10]
BWT [16]
Wavelet
Packet [11]
Proposed
0 dB
PESQ 2.39 2.19 2.25 2.22 2.27 2.36
segSNR 4.78 5.43 5.5 2.86 6.03 5.5
5 dB
PESQ 2.59 2.54 2.64 2.26 2.71 2.69
segSNR 3.48 5.12 4.96 3.69 5.52 4.89
10 dB
PESQ 2.90 2.89 2.92 2.70 2.88 2.95
segSNR 1.45 4.12 4.64 4.78 4.90 4.77
Table 5. Comparative performance for train noise
Input SNR
Parameter
Name
OM-LSA
[13]
SMPO [11]
MMSE-SPP
[10]
BWT [16]
Wavelet
Packet [11]
Proposed
0 dB
PESQ 2.15 2.17 2.07 2.08 2.26 2.29
segSNR 4.85 3.84 4.77 3.45 4.90 5.50
5 dB
PESQ 2.51 2.22 2.48 2.28 2.53 2.59
segSNR 4.5 2.43 4.28 3.89 4.06 4.79
10 dB
PESQ 2.75 2.67 2.76 2.59 2.89 2.94
segSNR 4.06 1.07 3.71 3.5 3.64 4.47
The Figure 7, Figure 8, Figure 9, Figure 10 shows that comparative analysis representation of
proposed approach in terms of PESQ for babble, airport, car and train noises for 0 dB, 5 dB and 10 dB
signals, considering the five states of art techniques. The PESQ measurement and improvement in the Table
2, 3, 4 and 5 shows that proposed method develops the quality of the enhanced signal at a rate of 7.76%
average improvement for 0 dB, 2.65% improvement for 5 dB and finally 3.78% improvement is marked at 10
dB SNR signals.
Figure 7. PESQ performance analysis for babble noise

71
Figure 8. PESQ performance analysis for airport noise
Figure 9. PESQ performance analysis for car noise Figure 10. PESQ performance analysis for train
noise
4. CONCLUSION
The objective intelligibility measured for various voiced and unvoiced signals is found to be better
in the proposed method. Wiener filtering is implemented along with spectral coefficient, and probability
distribution models, to improve the performance of speech quality in terms of PESQ and segSNR. While
extracting the noise characteristics phase coefficients are preserved and reused after the reconstruction. It
performs better in case of low SNR where noise power is calculated locally. Furthermore, SPP is also used to
make a soft-decision in detecting the noise statistics in the absence of the speech and MMSE based noise
tracking is used in the speech presence. This gives better noise estimation and improves the performance at
high SNR. Finally, recursive smoothing is applied resulting in the efficient single-channel speech
enhancement. The speech enhancement using adaptive filtering may reduce the cepstral smoothing, also echo
cancelation and improve the results further. The results indicate that the proposed work is improving the
speech quality at 0 dB, 5 dB and 10 dB and stands mostly good, and comparative in some particular cases.
Speech intelligibility is also good according to the Mean Opinion Score (MOS) of the subjects. Also, it
improves the quality of the lower energy concentrated signals efficiently when compared to the other state of
art techniques.

 ISSN: 2252-8776
72
REFERENCES
[1] R. E. Kalman, “A New Approach to Linear Filtering and Prediction Problems,” Journal of Basic Engineering,
vol.82, no.1, pp. 35-45, 1960.
[2] Boll, S. F, “Suppression of acoustic noise in speech using spectral subtraction,” IEEE Transactions on Acoustics,
Speech, and Signal Processing, pp. 113–120, 1979.
[3] M. Berouti, R. Schwartz, and J. Makhoul, “Enhancement of speech corrupted by acoustic noise,” in the Proceedings
of IEEE International Conference on Acoustics, Speech and Signal Processing., pp. 208–211, 1979.
[4] K. T. Talale, and S.T. Gandhe, “Speech Compression using ADPCM,” IJCA - International Journal of Computer
Applications In the Proceedings on International Conference in Computational Intelligence, no. 8, 2012.
[5] L. R. Rabiner, and R. W. Schafer, “Digital Processing of Speech Signal,” The Journal of the Acoustical Society of
America, vol. 67, no. 4, pp.1406-1407, 1980.
[6] H. J. M. Steeneken and T. Houtgast, “A physical method for measuring speech-transmission quality,” Journal of the
Acoustical Society of America, vol. 67, no. 1, pp. 318–326, 1980.
[7] T. Gerkmann and R. C. Hendriks, “Noise Power Estimation Based on The Probability of Speech Presence,” IEEE
Workshop on Applications of Signal Processing to Audio and Acoustics, 2011.
[8] Park, Jeong-Sik, Gil-Jin Jang, and Ji-Hwan Kim. "Noise Reduction Scheme for Speech Recognition in Mobile
Devices," Information Technology Convergence. Springer, Dordrecht, pp. 949-955, 2013.
[9] Zhao, Huan, et al. "A new soft masking method for speech enhancement in the frequency domain." Elektronika ir
Elektrotechnika, vol. 20, no. 2, pp. 58-63, 2014.
[10] J. Jensen, R. C. Hendriks, and T. Gerkman, “DFT domain based single-microphone noise reduction for speech
enhancement,” A survey of the state-of-the-art, 2013.
[11] P. Sun, Jun Qin, “Speech enhancement via two-stage dual tree complex wavelet packet transform with a speech
presence probability estimator”, The Journal of Acoustic Society of America, vol. 141, no. 2, pp. 808-817, Jan 2017.
[12] Young-Seok Choi, “A New Subband Adaptive Filtering Algorithm for Sparse System Identification with Impulsive
Noise”, Journal of Applied Mathematics, vol. 2014.
[13] Tran, Tien Dung, Quoc Cuong Nguyen, and Dang Khoa Nguyen. "Speech enhancement using modified IMCRA and
OMLSA methods," International Conference on Communications and Electronics 2010., IEEE, 2010.
[14] I. Cohen, “Noise spectrum estimation in adverse environments: Improved minima controlled recursive averaging,”
IEEE Transaction in Speech Audio Processing, vol. 11, pp. 466-475, 2003.
[15] Zhao, Huan, et al. "A new soft masking method for speech enhancement in the frequency domain." Elektronika ir
Elektrotechnika, vol. 20. no. 2, pp. 58-63, 2014.
[16] Sonu Bala, and Mohammad Arif, “Performance comparison of discrete transforms on speech compressed sensing,”
In the Proceedings of IEEE International Conference on Computational Intelligence & Communication Technology,
pp.632-637, 2015.

01 8445 speech enhancement

More Related Content

What's hot (19)

Similar to 01 8445 speech enhancement (20)

More from IAESIJEECS (20)

Recently uploaded (20)

01 8445 speech enhancement