We use cookies to give you the best experience possible. By continuing we’ll assume you’re on board with our cookie policy

See Pricing

What's Your Topic?

Hire a Professional Writer Now

The input space is limited by 250 symbols

What's Your Deadline?

Choose 3 Hours or More.
2/4 steps

How Many Pages?

3/4 steps

Sign Up and See Pricing

"You must agree to out terms of services and privacy policy"
Get Offer

Audio Watermarking Sample

Hire a Professional Writer Now

The input space is limited by 250 symbols

Deadline:2 days left
"You must agree to out terms of services and privacy policy"
Write my paper

1. Introduction

With the rapid development of the address. sound. image. and video compaction methods. presently it is non a hard undertaking to distribute digital multimedia over Internet. This makes the protections of digital rational belongings rights and content hallmarks have been a serious job. Hence the engineering of digital watermarking is received a big trade of attending. Generally. digital watermarking techniques are based on either spread spectrum methods or altering the least important spots of selected coefficients of a certain signal transform.

Don't use plagiarized sources. Get Your Custom Essay on
Audio Watermarking Sample
Just from $13,9/Page
Get custom paper

For address watermarking. to guarantee the embedded water line is unperceivable. the audio marker phenomenon is considered together with these conventional techniques.

In add-on. a address watermarking system should be robust to assorted speech compaction operations. The development of address watermarking algorithms. hence. involves a tradeoff among speech fidelity. hardiness. and watermark form implanting rate specifications. The address watermarking techniques normally embed speech water line in unneeded parts of speech signal. or in human insensitiveness auditory parts. Some of address watermarking methods will alter an interval to implant water line.

However. this sort of method has a drawback that is the unavoidably debasement of hardiness.

In the other methods. the water lines are embedded by the usage of forgery human address. It is unfortunate that such type of method besides has the defect of weak hardiness particularly when the forgery human address is destroyed. The deformation of the forgery human address will besides take to the harm of the water line

Fig 1. 1: Block of General watermarking Scheme

Therefore. we can specify watermarking systems as systems in which the hidden message is related to the host signal and non-watermarking systems in which the message is unrelated to the host signal. On the other manus. systems for implanting messages into host signals can be divided into steganographic systems. in which the being of the message is unbroken secret. and non-steganographic systems. in which the presence of the embedded message does non hold to be secret.

Audio watermarking algorithms are characterized by five indispensable belongingss. viz. perceptual transparence. watermark spot rate. hardiness. blind/informed watermark sensing. and security

Perceptual transparence

In most of the applications. the watermark-embedding algorithm has to infix extra informations without impacting the perceptual quality of the sound host signal. The fidelity of the watermarking algorithm is normally defined as a perceptual similarity between the original and watermarked audio sequence. However. the quality of the watermarked sound is normally degraded. either deliberately by an antagonist or accidentally in the transmittal procedure. before a individual perceives it. In that instance. it is more equal to specify the fidelity of a watermarking algorithm as a perceptual similarity between the watermarked sound and the original host sound at the point at which they are presented to a consumer.

Watermark spot rate

The spot rate of the embedded water line is the figure of the embedded spots within a unit of clip and is normally given in spots per second ( bits per second ) . Some audio watermarking applications. such as transcript control. necessitate the interpolation of a consecutive figure or writer ID. with the mean spot rate of up to 0. 5 bits per second. For a broadcast monitoring water line. the spot rate is higher. caused by the necessity of the embedding of an ID signature of a commercial within the first second at the start of the broadcast cartridge holder. with an mean spot rate up to 15 bits per second. In some pictured applications. e. g. concealing address in sound or compressed audio watercourse in sound. algorithms have to be able to implant water lines with the spot rate that is a important fraction of the host audio spot rate. up to 150 kbps.


The hardiness of the algorithm is defined as an ability of the water line sensor to pull out the embedded water line after common signal processing uses. A elaborate overview of hardiness trials is given in Chapter 3. Applications normally require hardiness in the presence of a predefined set of signal processing alterations. so that water line can be faithfully extracted at the sensing side. For illustration. in wireless broadcast monitoring. embedded water line demand merely to last deformations caused by the transmittal procedure. including dynamic compaction and low base on balls filtering. because the water line sensing is done straight from the broadcast signal. On the other manus. in some algorithms hardiness is wholly unwanted and those algorithms are labeled delicate sound watermarking algorithms.

Blind or informed watermark sensing

In some applications. a sensing algorithm may utilize the original host sound to pull out water line from the watermarked audio sequence ( informed sensing ) . It frequently significantly improves the sensor public presentation. in that the original sound can be subtracted from the watermarked transcript. ensuing in the water line sequence entirely. However. if sensing algorithm does non hold entree to the original sound ( blind sensing ) and this inability well decreases the sum of informations that can be hidden in the host signal. The complete procedure of implanting and pull outing of the water line is modeled as a communications channel where water line is distorted due to the presence of strong intervention and channel effects. A strong intervention is caused by the presence of the host sound. and channel effects correspond to signal processing operations.


Watermark algorithm must be secure in the sense that an antagonist must non be able to observe the presence of embedded information. allow entirely take the embedded information. The security of water line procedure is interpreted in the same manner as the security of encoding techniques and it can non be broken unless the authorized user has entree to a secret key that controls watermark implanting. An unauthorised user should be unable to pull out the informations in a sensible sum of clip even if he knows that the host signal contains a water line and is familiar with the exact water line implanting algorithm. Security demands vary with application and the most rigorous are in screen communications applications. and. in some instances. information is encrypted prior to implanting into host sound.

Theory: –

The cardinal procedure in each watermarking system can be modeled as a signifier of communicating where a message is transmitted from water line embedder to the water line receiving system. The procedure of watermarking is viewed as a transmittal channel through which the water line message is being sent. with the host signal being a portion of that channel. In Figure 2. a general function of a watermarking system into a communications theoretical account is given. After the water line is embedded. the watermarked work is normally distorted after water line onslaughts. The deformations of the watermarked signal are. likewise to the informations communications theoretical account. modeled as linear noise.

Fig 1. 2: Basic Watermarking system tantamount to a communicating system

In this undertaking. signal processing methods are used for water line embedding and extracting procedures. derivation of perceptual thresholds. transforms of signals to different signal spheres ( e. g. Fourier sphere. ripple sphere ) . filtering and spectral analysis. Communication rules and theoretical accounts are used for channel noise mold. different ways of signaling the water line ( e. g. a direct sequence spread spectrum method. frequence skiping method ) derivation of optimized sensing method ( e. g. matched filtrating ) and rating of overall sensing public presentation of the algorithm ( bit mistake rate. normalized correlativity value at sensing ) . The basic information theory rules are used for the computation of the perceptual information of an audio sequence. channel capacity bounds of a water line channel and during design of an optimum channel coding method.

During transmittal and response signals are frequently corrupted by noise. which can do terrible jobs for downstream processing and user perceptual experience. It is good known that to call off the noise constituent nowadays in the standard signal utilizing adaptative signal processing technique. a mention signal is needed. which is extremely correlated to the noise. Since the noise gets added in the channel and is wholly random. hence there is no agency of making a correlative noise. at the having terminal. Lone manner possible is to somehow pull out the noise. from the standard signal. itself. as merely the standard signal can state the narrative of the noise added to it. Therefore an machine-controlled agencies of taking the noise would be an priceless first phase for many signal-processing undertakings. Denoising has long been a focal point of research and yet there ever remains room for betterment.

Simple methods originally employed the usage of time-domain filtering of the corrupted signal. nevertheless. this is merely successful when taking high frequence noise from low frequence signals and does non supply satisfactory consequences under existent universe conditions. To better public presentation. modern algorithms filter signals in some transform sphere such as omega for Fourier. Over the past two decennaries. a bustle of activity has involved the usage of the ripple transform after the community recognized the possibility that this could be used as a superior option to Fourier analysis. Numerous signal and image processing techniques have since been developed to leverage the power of ripples. These techniques include the distinct ripple transform. ripple package analysis. and most late. the lifting strategy.

2. Address Processing

Speech Production: -Address is produced when air is forced from the lungs through the vocal cords and along the vocal piece of land. The vocal piece of land extends from the gap in the vocal cords ( called the glottis ) to the oral cavity. and in an mean adult male is about 17 centimeters long. It introduces short-run correlativities ( of the order of 1ms ) into the speech signal. and can be thought of as a filter with wide resonances called formants. The frequences of theseformants are controlled by changing the form of the piece of land. for illustration by traveling the place of the lingua. An of import portion of many address codec is the mold of the vocal piece of land as a short-run filter. As the form of the vocal piece of land varies comparatively easy. the transportation map of its mold filter needs to be updated merely comparatively infrequently ( typically every 20 MS or so ) .

From the proficient. signal-oriented point of position. the production of address is widely described as a two-level procedure. In the first phase the sound is initiated and in the 2nd phase it is filtered on the 2nd degree. This differentiation between stages has its beginning in the source-filter theoretical account of speech production.

Fig 3: Beginning Filter Model of Speech Production

The basic premise of the theoretical account is that the beginning signal produced at the glottal degree is linearly filtered through the vocal piece of land. The ensuing sound is emitted to the environing air through radiation burden ( lips ) . The theoretical account assumes that beginning and filter are independent of each other. Although recent findings show some interaction between the vocal piece of land and a glottal beginning ( Rothenberg 1981 ; Fant 1986 ) . Fant’s theory of address production is still used as a model for the description of the human voice. particularly every bit far as the articulation of vowels is concerned.

Address Processing: -The term address treating fundamentally refers to the scientific subject refering the analysis and processing of address signals in order to accomplish the best benefit in assorted practical scenarios. The field of address processing is. at present. undergoing a rapid growing in footings of both public presentation and applications. The progresss being made in the field of microelectronics. calculation and algorithm design stimulate this. Nevertheless. address processing still covers an highly wide country. which relates to the undermentioned three technology applications:

• Speech Coding and transmittal that is chiefly concerned with man-to adult male voice communicating. • Speech Synthesis which deals with machine-to-man communications. • Speech Recognition associating to man-to machine communicating.

Address Cryptography: -Speech cryptography or compaction is the field concerned with compact digital representations of address signals for the intent of efficient transmittal or storage. The cardinal aim is to stand for a signal with a minimal figure of spots while keeping perceptual quality. Current applications for address and audio cryptography algorithms include cellular and personal communications webs ( PCNs ) . teleconferencing. desktop multi-media systems. and unafraid communications.

The Discrete Wavelet Transform: -The Discrete Wavelet Transform ( DWT ) involves taking graduated tables and places based on powers of two’s. So called dyadic graduated tables and places. The female parent ripple is rescaled or dilated by powers of two and translated by whole numbers. Specifically. a map degree Fahrenheit ( T ) L2 ( R ) ( defines infinite of square built-in maps ) can be represented as

( 3. 4 )

The map ? ( T ) is known as the female parent ripple. while ? ( T ) is known as the grading Function. The set of maps

Where Z is the set of whole numbers is an ortho- normal footing for L2 ( R ) . The Numberss a ( L. K ) are known as the estimate coefficients at scale L. while vitamin D ( j. K ) are known as the item coefficients at scale J.

The estimate and item coefficients can be expressed as:

To supply some apprehension of the above coefficients see a projection Florida ( T ) of the map degree Fahrenheit ( T ) that provides the best estimate ( in the sense of minimal mistake energy ) to f ( T ) at a scale l. This projection can be constructed from the coefficients a ( L. k ) . utilizing the equation ( 3. 5 )

As the graduated table cubic decimeter decreases. the estimate becomes finer. meeting to f ( T ) as cubic decimeter > 0. The difference between the estimate at scale cubic decimeter + 1 and that at l. fl+1 ( T ) – Florida ( T ) . is Wholly described by the coefficients vitamin D ( j. K ) utilizing the equation

( 3. 6 )

Using these dealingss. given a ( L. k ) and { vitamin D ( j. K ) | j ? L } . it is clear that we can construct the estimate at any graduated table. Hence. the ripple transform breaks the signal up into a harsh estimate Florida ( T ) ( given a ( L. k ) ) and a figure of beds of item { fj+1 ( T ) -fj ( T ) | J & lt ; L } ( given by { vitamin D ( j. K ) | j ? L } ) . As each bed of item is added. the estimate at the following finer graduated table is achieved.

Disappearing MomentsThe figure of disappearing minutes of a ripple indicates the smoothness of the ripple map every bit good as the two-dimensionality of the frequence response of the ripple filters ( filters used to calculate the DWT ) . Typically a ripple with p disappearing minutes satisfies the undermentioned equation or equivalently.

For the representation of smooth signals. a higher figure of disappearing minutes leads to a faster decay rate of ripple coefficients. Therefore. ripples with a high figure of disappearing minutes lead to a more compact signal representation and are therefore utile in codingapplications. However. in general. the length of the filters increases with the figure of disappearing minutes and the complexness of calculating the DWT coefficients increases with the size of the ripple filters.

The Fast Wavelet Transform AlgorithmThe Discrete Wavelet Transform ( DWT ) coefficients can be computed by utilizing Mallat’s Fast Wavelet Transform algorithm. This algorithm is sometimes referred to as the stereophonic sub-band programmer and involves filtrating the input signal based on the ripple map used. Implementation Using Filters To explicate the execution of the Fast Wavelet Transform algorithm see the undermentioned equations:

( 3. 9 )The first equation is known as the twin-scale relation ( or the dilation equation ) and defines the grading map ? . The following equation expresses the ripple ? in footings of the grading map ? . The 3rd equation is the status required for the ripple to be Orthogonal to the grading map and its translates

The coefficients c ( K ) or { c0. . . . c2N-1 } in the above equations represent the impulse response coefficients for a low base on balls filter of length 2N. with a amount of 1 and a norm of1/2 The high base on balls filter is obtained from the low base on balls filter utilizing the relationship g ( ) K degree Celsius ( K ) K = ?1 1? . where K varies over the scope ( 1. ( 2N. 1 ) ) to 1.

Equation 2. 7 shows that the grading map is basically a low base on balls filter and is used to specify the estimates. The ripple map defined by equation 2. 8 is a high base on balls filter and defines the inside informations. Get downing with a distinct input signal vector s. the first phase of the FWT algorithm decomposes the signal into two sets of coefficients. These are the estimate coefficients cA1 ( low frequence information ) and the item coefficients cD1 ( high frequence information ) . as shown in the figure below.

Fig 3. 3: Block diagram of DWT

The coefficient vectors are obtained by convoluting s with the low-pass filter Lo_D for Approximation and with the high-pass filter Hi_D for inside informations. This filtering operation is Then followed by dyadic decimation or down trying by a factor of 2. Mathematically the stereophonic filtering of the distinct signal s is represented by the looks:

( 3. 10 )

These equations implement a whirl plus down sampling by a factor 2 and give the forward fast ripple transform. If the length of each filter is equal to 2N and the length of the original signal s is equal to n. so the corresponding lengths of the coefficients cA1 and cD1 are given by the expression:

( 3. 11 )

This shows that the entire length of the ripple coefficients is ever somewhat greater than the length of the original signal due to the filtering procedure used.

Multilevel DecompositionThe decomposition procedure can be iterated. with consecutive estimates being decomposed in bend. so that one signal is broken down into many lower declaration Components. This is called the ripple decomposition tree.

Fig 3. 4: Wavelet decomposition treeThe ripple decomposition of the signal s analysed at degree J has the undermentioned construction [ cAj. cDj. … . cD1 ] .Looking at a signals wavelet decomposition tree can uncover valuable information.

Since the analysis procedure is iterative. in theory it can be continued indefinitely. In world. the decomposition can merely continue until the vector consists of a individual sample. Normally. nevertheless there is small or no advantage gained in break uping a signal beyond a certain degree. The choice of the optimum decomposition degree in the hierarchy depends on the nature of the signal being analysed or some other suited standard. such as the low-pass filter cut-off.

Signal Reconstruction

The original signal can be reconstructed or synthesised utilizing the reverse distinct ripple transform ( IDWT ) . The synthesis starts with the estimate and item coefficients cAj and cDj. and so reconstructs cAj-1 by up sampling and filtrating with the Reconstruction filters.

Fig 3. 5: Block diagram of IDWT

The Reconstruction filters are designed in such a manner to call off out the effects of aliasing introduced in the ripple decomposition stage. The Reconstruction filters ( Lo_R and Hi_R ) together with the low and high base on balls decomposition filters. organize a system known as quadrature mirror filters ( QMF ) .


Speech H2O marker:The address H2O taging agencies embed a digital information ( address ) into the other speech signal ( . wav ) or take the signal constituents of coveted signal is called speech H2O taging Here I have consider two speech signal like

1 ) Select the coveted address signal ( . wav ) . read coveted moving ridge length and play the selected desired speech signal

2 ) Select the embedded address signal ( . wav ) . read and play the selected embedded speech signal

3 ) Select the coveted address signal ( . wav ) . read selected coveted address signal

4 ) Than above signals applied distinct moving ridge let transform with name of the ripple “Haar” . Because we need required desired processing

5 ) Due to speech H2O marker: the coveted signals one by one processing continues

Here I have used cat map.

6 ) Here H2O taging consequences playing

7 ) SWPR: base for speech H2O taging signal drama. due to entering

8 ) SWRP: base for speech H2O taging signal recorded and playing

Cite this Audio Watermarking Sample

Audio Watermarking Sample. (2017, Oct 25). Retrieved from https://graduateway.com/audio-watermarking-essay-sample-essay/

Show less
  • Use multiple resourses when assembling your essay
  • Get help form professional writers when not sure you can do it yourself
  • Use Plagiarism Checker to double check your essay
  • Do not copy and paste free to download essays
Get plagiarism free essay

Search for essay samples now

Haven't found the Essay You Want?

Get my paper now

For Only $13.90/page