IOSR Journal of Engineering (IOSR JEN) www.iosrjen.

org
ISSN (e): 2250-3021, ISSN (p): 2278-8719
PP 32-36

Audio Compression Using Wavelet Transform

Swapnil T. Dumbre 1, Neeta B. Bankhele 2
1
(Department of E&TC Samarth College of Engineering Belhe, Pune, India
2
(Department of E&TC Sharadchandra College of Engineering dumberwadi, Pune ,India

Abstract: Audio Compression is one of the basic technologies of the modern telecommunication age.
Compression is the technique to convert high input data stream into smaller size. Audio coding gives us the
digital form of audio with as few bits as possible and also maintains the quality. the reduction in bit rates
conserve bandwidth . Audio coding is used in various applications such as digital broadcasting ,high quality
audio for satellite transmission , internet audio or music database where the high quality audio signals bit rate
is reduced without compromising the quality of the signal.
The technology proposed to achieve the design and implementation of audio compression using discrete wavelet
transform technique. The efficiency performance of the audio encoding methods has been measured using
compression ratio as well as peak signal to noise (PSNR) ratio, SNR.

I. Introduction
The revolution in computer history had invariably led to the demand of quality audio data. But the data
rates associated with the uncompressed audio signal is massive. The product of the sampling rate and number of
bits is known as the bit rate. The subtraction of the information rate and bit rate of the signal is known as
redundancy. The audio compression
Technique works to reduce this redundancy without affecting the quality of the audio signal it has
found many applications in various areas such as multimedia signal coding, high fidelity audio for radio
broadcasting, audio transmission for HDTV, audio data transmission /sharing through internet etc.
Sheetal D. Gunjal, Rajeshee D. Raut, 2015. Traditional Psychoacoustic model and daubechies
wavelets for enhanced speech coder performance described the dependencies of Compression ratio, SNR and
the decomposition level it shows increase in the compression ratio value limited by the SNR value.[1] Oathman
O.Khalifa, Sering Habib Harding and Aisha-Hassan described the method for audio compression the signal is
compressed using wavelet and Reconstructed signals are compared using factors like Signal to Noise Ratio
(SNR), Peak Signal to Noise Ratio (PSNR), Normalized Root Mean Square Error (NRMSE), Compression
Ratio for different levels of wavelet .[6] P. Srinivasan and L. H. Jamieson. “High Quality Audio Compression
Using an Adaptive Wavelet Packet Decomposition and Psychoacoustic Modelling described Wavelet packet-
based compression scheme suitable for high-quality audio transfers over the internet or storage.[5]
To reduce the coded bit rate there are basically two types of techniques used in the first type some form
of digital encoding is applied on the basis of statistical redundancy it is a lossless audio coding in which the
original audio signal is remains unharmed and can be totally recovered. In the second type some sort of signal
processing is used so that sorting of unwanted and information signal components can be done the recovered
audio signal and the original signal is not identical therefore we can say this technique is lossy technique.

1.1 Types of Compression

a) Lossless compression:-
As their name implies, involve no loss of information. if data have been losslessly compressed, the original data
can be recovered exactly from the compressed data lossless compression is generally used for applications that
cannot tolerate any difference between the original and reconstructed data.
b) Lossy compression:-
Lossy compression techniques involve some loss of information, and data that have been compressed using
lossy techniques generally cannot be recovered or reconstructed exactly. In return for accepting this distortion in
the reconstruction, we can generally obtain much higher compression ratios than is possible with lossless
compression.

II. Techniques for Audio compression

Audio compression is classified into three methods
A. Waveform coding
B. Parametric coding
C. Transform coding
National Conference on “Recent Innovations in Engineering and Technology” MOMENTUM-19 32 | Page
Sharadchandra Pawar College of Engineering, Dumbarwadi, Tal-Junnar, Dist-Pune-410504
 Audio Compression Using Wavelet Transform

i) Fast Fourier Transform (FFT)

ii) Discrete Cosine Transform (DCT)
iii) Continuous Wavelet transform(CWT)
iv) Discrete Wavelet transform (DWT)

2.1 Introduction to wavelet:

The fundamental idea behind wavelets is to analyse according to scale. The wavelet analysis procedure
is to adopt a wavelet prototype function called an analysing wavelet or mother wavelet. Any signal can then be
represented by translated and scaled versions of the mother wavelet. Wavelet analysis is capable of revealing
aspects of data that other signal analysis techniques such as Fourier analysis miss, aspects like trends,
breakdown points, discontinuities in higher derivatives, and self-similarity. Furthermore, because it affords a
different view of data than those presented by traditional techniques, it can compress or de-noise a signal
without appreciable degradation. The different types of wavelet families are like Daubechies, haar, symmlet etc.

Figure 1: Different wavelet families

2.2Audio compression using wavelet:

Transform means converting time domain signal into the frequency domain. Transform coding is the
type of data compression for natural data like audio signal. In this, the knowledge of the application is used to
choose the information to discard, thereby lowering its bandwidth.

National Conference on “Recent Innovations in Engineering and Technology” MOMENTUM-19 33 | Page

Sharadchandra Pawar College of Engineering, Dumbarwadi, Tal-Junnar, Dist-Pune-410504
 Audio Compression Using Wavelet Transform

A) Transform Technique :
Discrete wavelet transform is used for the audio signal compression. Wavelet transform is very suitable for
audio compression. DWT uses multiresolution technique to analyze different frequencies.
B) Quantization and coding of transform coefficients :
If the amount of information conveyed by each coefficient is different, it makes sense to assign differing
numbers of bits to the different coefficients. The Process of quantization is to convert the discrete time
continuous amplitude into discrete time discrete amplitude. This is done by rounding off each sample to the
nearest quantization level. Each discrete time, discrete amplitude is further represented by finite number of
digits using a coder.
C) Encoding :
In Encoding techniques reduce the number of coefficients by removing the redundant data. The compressed
audio signal can be reconstructed to form the original signal by Decoding followed by de-quantization and
then performing inverse transform Methods.

III Result
3.1 Compression Ratio:
The compression ratio (CR) is defined as the ratio of the length of original signal to the length of the
compressed signal.
Length of original signal
Compression Ratio =
Length of compressed signal

3.2 PSNR (Packet Signal to Noise Ratio):

It is Defined as the ratio between the maximum possible power of a signal and the power of corrupting
noise that affects the fidelity of its representation. Because many signals have a very wide dynamic range,
PSNR is usually expressed in terms of the logarithmic decibel scale.

3.3 SNR (Signal to Noise Ratio):

σx2 is the mean square of the speech signal and σe2 is the mean square difference between the
original and reconstructed signals.

Figure 3: Original & Compressed audio signal

National Conference on “Recent Innovations in Engineering and Technology” MOMENTUM-19 34 | Page
Sharadchandra Pawar College of Engineering, Dumbarwadi, Tal-Junnar, Dist-Pune-410504
 Audio Compression Using Wavelet Transform

level Parameter DB02 DB 04 DB06 DB08 DB10

C.R. 10 11 9 7 9
1
PSNR 33 35 36 36 37
C.R. 22 18 18 18 19
2
PSNR 27 28 28 28 28
C.R. 38 34 32 31 31
3
PSNR 24 25 25 25 25
C.R. 60 55 54 51 54
4
PSNR 22 23 23 23 23
C.R. 68 62 60 57 58
5
PSNR 22 22 22 22 22

Table -1 Performance parameters in the proposed coder

Level of Daubenchies Wavelet Vs PSNR

40
35
30 PSNR
25
PSNR
PSNR

20
15 PSNR
10 PSNR
5 PSNR
0
DB02 DB04 DB06 DB08 DB10

Fig.4 Daubenchies Wavelet Family Vs PSNR

Level of Daubenchies Wavelet Vs CR

80
70
Compression Ratio

60 level 1 C.R.
50
level 2 C.R.
40
30 level 3 C.R.
20 level 4 C.R.
10
level 5 C.R.
0
DB02 DB04 DB06 DB08 DB10

Fig.5 Daubenchies Wavelet Family Vs Compression Ratio

IV. Conclusion
The selection of the Daubechies wavelet family with DWT yielded comparable improvement in the
performance parameters with a good quality reconstruction of the speech signal. The compression factor
improves at the cost of the SNR with progressive levels. At levels 3, 4 and 5 the variation in CF and SNR is
much more consistent. One can select level 3 for good performance, with a moderate number of filter banks.
Adaptive filter banks can be in combination with a model for more effective coding. In addition, using wavelets
the compression ratio can be easily varied, while most other compression techniques have fixed compression
ratios.

National Conference on “Recent Innovations in Engineering and Technology” MOMENTUM-19 35 | Page

Sharadchandra Pawar College of Engineering, Dumbarwadi, Tal-Junnar, Dist-Pune-410504
 Audio Compression Using Wavelet Transform

The discrete wavelet transform performs very well in the compression of audio signals. For real time
audio processing however, its performance is not as good. Therefore for real time audio coding it is
recommended to use a wavelet with a small number of vanishing moments at level 5decomposition or less.

References
[1]. Sheetal D. Gunjal, Rajeshee D. Raut, 2015. Traditional Psychoacoustic model and daubechies wavelets for enhanced speech
coder performance: International Journal of Technology (2015) 2: 190-197 ISSN 2086-9614
[2]. Sheetal D. Gunjal, Rajeshee D. Raut, 2012. Advance Source Coding Techniques for Audio/Speech Signal: A Survey,
International Journal for Computer Technology and Applications, Volume 3(4), pp. 1335-1342
[3]. M.V. Patil, Apporva Gupta, Ankita Verma and Shikhar Salil, “Audio and Speech Compression Using DCT and DWT
Techniques”, IJIRSET International Journal of Innovative Research in Science, Engineering and Technology, Vol.2 ,Issue 5,
May 2013.
[4]. HarmanpreetKaur and RamanpreetKaur, “Speech compression and decompression using DCT and DWT”, International Journal
Computer Technology &Applications,Vol 3 (4), 1501-1503 IJCTA | July-August 2012
[5]. P. Srinivasan and L. H. Jamieson. “High Quality Audio Compression Using an Adaptive
Wavelet Packet Decomposition and Psychoacoustic Modelling”, IEEE Transactions on Signal Processing, Vol. 46, No. 4, April
1998
[6]. Othman O. Khalifa, Sering Habib Harding & Aisha-Hassan A. Hashim “Compressionusing Wavelet Transform” in Signal
Processing: An International Journal, Volume (2):
[7]. Khalid sayood . "Introduction to Data Compression " 4th Edition. Elsevier
[8]. S.G. Mallat. "A Wavelet Tour of Signal Processing." 2nd Edition. Academic Press, 1999. ISBN 0-12-466606-X

National Conference on “Recent Innovations in Engineering and Technology” MOMENTUM-19 36 | Page

Sharadchandra Pawar College of Engineering, Dumbarwadi, Tal-Junnar, Dist-Pune-410504