Title of Invention

METHOD FOR DETERMINING IF ONE AUDIO SIGNAL IS DERIVED FROM ANOTHER AUDIO SIGNAL OR IF, TWO AUDIO SIGNALS ARE DERIVED FROM THE SAME AUDIO SIGNAL USING CHARACTERIZATIONS BASED ON AUDITORY EVENTS

Abstract A method for determining if one audio signal is derived from another audio signal or if two audio signals are derived from the same audio signal compares reduced-information characterizations of said audio signals, wherein said characterizations are based on auditory scene analysis. The comparison removes from the characterisations or minimizes in the characterisations the effect of temporal shift or delay on the audio signals (5-1), calculates a measure of similarity (5-2), and compares the measure of similarity against a threshold. In one alternative, the effect of temporal shift or delay is removed or minimized by cross- correlating the two characterizations. In another alternative, the effect of temporal shift or delay is removed or minimized by transforming the characterizations into a domain that is independent of temporal delay effects, such as the frequency domain. In both cases, a measure of similarity is calculated by calculating a coefficient of correlation.
Full Text METHOD FOR DETERMINING IF ONE AUDIO SIGNAL IS DERIVED FROM
ANOTHER AUDIO SIGNAL OR IF, TWO AUDIO SIGNALS ARE DERIVED
FROM THE SAME AUDIO SIGNAL USING CHARACTERIZATIONS BASED
ON AUDITORY EVENTS
DESCRIPTION
Comparing Audio Using Characterizations Based on Auditory Events
TECHNICAL FIELD
The invention relates to audio signals. More particularly, the invention relates
to characterizing audio signals and using characterizations to determine if one audio
signal is derived from another audio signal or if two audio signals are derived from
the same audio signal.
BACKGROUND ART
The division of sounds into units perceived as separate is sometimes referred
to as "auditory event analysis" or "auditory scene analysis" ("ASA"). An extensive
discussion of auditory scene analysis is set forth by Albert S. Bregman in his book
Auditory Scene Analysis - The Perceptual Organization of Sound, Massachusetts
Institute of Technology, 1991, Fourth printing, 2001, Second MIT Press paperback
edition. In addition, United States Patent 6,002,776 to Bhadkamkar, et al, December
14, 1999 cites publications dating back to 1976 as "prior art work related to sound
separation by auditory scene analysis." However, the Bhadkamkar, et al patent
discourages the practical use of auditory scene analysis, concluding that
" [t]echniques involving auditoiy scene analysis, although interesting from a scientific
point of view as models of human auditoiy processing, are currently far too
computationally demanding and specialized to be considered practical techniques for
sound separation until fundamental progress is made."
Bregman notes in one passage that "[w]e hear discrete units when the sound
changes abruptly in timbre, pitch, loudness, or (to a lesser extent) location in space."
(Auditory Scene Analysis - The Perceptual Organization of Sound, supra at page
469). Bregman also discusses the perception of multiple simultaneous sound streams
when, for example, they are separated in frequency.
There are many different methods for extracting characteristics or features
from audio. Provided the features or characteristics are suitably defined, their
extraction can be performed using automated processes. For example "ISO/EEC JTC

1/SC 29/WG 11" (MPEG) is currently standardizing a variety of audio descriptors as
part of the MPEG-7 standard. A common shortcoming of such methods is that they
ignore ASA. Such methods seek to measure, periodically, certain "classical" signal
processing parameters such as pitch, amplitude, power, harmonic structure and
spectral flatness. Such parameters, while providing useful information, do not
analyze and characterize audio signals into elements perceived as separate according
to human cognition.
Auditory scene analysis attempts to characterize audio signals in a manner
similar to human perception by identifying elements that are separate according to
human cognition. By developing such methods, one can implement automated
processes that accurately perform tasks that heretofore would have required human
assistance.
The identification of separately perceived elements would allow the unique
identification of an audio signal using substantially less information than the full
signal itself. Compact and unique identifications based on auditoiy events may be
employed, for example, to identify a signal that is copied from another signal (or is
copied from the same original signal as another signal).
DISCLOSURE OF THE INVENTION
A method is described that generates a unique reduced-information
characterization of an audio signal that may be used to identify the audio signal. The
characterization may be considered a "signature" or "fingerprint" of the audio signal.
According to the present invention, an auditoiy scene analysis (ASA) is performed to
identify auditory events as the basis for characterizing an audio signal. Ideally, the
auditoiy scene analysis identifies auditoiy events that are most likely to be perceived
by a human listener even after the audio has undergone processing, such as low bit
rate coding or acoustic transmission through a loudspeaker. The audio signal may be
characterized by the boundary locations of auditoiy events and, optionally, by the
dominant frequency subband of each auditoiy event. The resulting information
pattern constitutes a compact audio fingerprint or signature that may be compared to

one or more other such audio fingerprints or signatures. A determination that at least
a portion of the respective signatures are the same (to a desired degree of confidence)
indicates that the related portions of the audio signals from which the respective
signatures were derived are the same or were derived from the same audio signal.
The auditory scene analysis method according to the present invention
provides a fast and accurate method of comparing two audio signals, particularly
music, by comparing signatures based on auditory event information. ASA extracts
information or features underlying the perception of similarity, in contrast to
traditional methods of feature extraction that extract features less fundamental to
perceiving similarities between audio signals (such as pitch amplitude, power, and
harmonic structure). The use of ASA improves the chance of finding similarity in
material that has undergone significant processing, such as low bit coding or acoustic
transmission through a loudspeaker.
Although in principle the invention may be practiced either in the analog or
digital domain (or some combination of the two), in practical embodiments of the
invention, audio signals are represented by samples in blocks of data and processing
is done in the digital domain.
Referring to FIG. 1 A, auditory scene analysis 2 is applied to an audio signal in
order to produce a "signature" or "fingerprint," related to that signal. In this case,
there are two audio signals of interest. They may be similar in that one may be
derived from the other or both may have been previously derived from the same
original signal, but this is not known in advance. Thus, auditoiy scene analysis is
applied to both signals. For simplicity, FIG. 1A shows only the application of ASA
to one signal. As shown in FIG. 1B, the signatures for the two audio signals,
Signature 1 and Signature 2, are applied to a correlator or con-elation function 4 that
generates a correlation score. A user may set a minimum correlation score as
providing a desired degree of confidence that at least a portion of the two signatures
are the same. In practice, the two signatures may be stored data. In one practical
application, one of the signatures may be derived, for example, from an unauthorized
copy of a musical work and the other signature may be one of a large number of

signatures in a database (each signature being derived from a copyright owner's
musical work) against which the unauthorized copy signature is compared until a
match, to a desired degree of confidence, if any, is obtained. This may be conducted
automatically by a machine, the details of which are beyond the scope of the present
invention.
Because the signatures are representative of the audio signals but are
substantially shorter (i.e., they are more compact or have fewer bits) than the audio
signals from which they were derived, the similarity of the two signatures (or lack
thereof) can be detennined much faster than it would take to determine the similarity
between the audio signals.
Further details of FIGS. 1A and 1B are set forth below.
In accordance with aspects of the present invention, a computationally
efficient process for dividing audio into temporal segments or "auditory events" that
tend to be perceived as separate is provided.
A powerful indicator of the beginning or end of a perceived auditory event is
believed to be a change in spectral content. In order to detect changes in timbre and
pitch (spectral content) and, as an ancillary result, certain changes in amplitude, the
audio event detection process according to an aspect of the present invention detects
changes in spectral composition with respect to time. Optionally, according to a
further aspect of the present invention, the process may also detect changes in
amplitude with respect to tune that would not be detected by detecting changes in
spectral composition with respect to time.
In its least computationally demanding implementation, the process divides
audio into time segments by analyzing the entire frequency band of the audio signal
(full bandwidth audio) or substantially the entire frequency band (in practical
implementations, band limiting filtering at the ends of the spectrum are often
employed) and giving the greatest weight to the loudest audio signal components.
This approach takes advantage of a psychoacoustic phenomenon in which at smaller
time scales (20 msec and less) the ear may tend to focus on a single auditoiy event at
a given tune. This implies that while multiple events may be occurring at the same

time, one component tends to be perceptually most prominent and may be processed
individually as though it were the only event taking place. Taking advantage of this
effect also allows the auditory event detection to scale with the complexity of the
audio being processed. For example, if the input audio signal being processed is a
solo instrument, the audio events that are identified will likely be the individual notes
being played. Similarly for an input voice signal, the individual components of
speech, the vowels and consonants for example, will likely be identified as individual
audio elements. As the complexity of the audio increases, such as music with a
drumbeat or multiple instruments and voice, the auditory event detection identifies
the most prominent (i.e., the loudest) audio element at any given moment.
Alternatively, the "most prominent" audio element may be detennined by taking
hearing threshold and frequency response into consideration.
Optionally, according to further aspects of the present invention, at the
expense of greater computational complexity, the process may also take into
consideration changes in spectral composition with respect to time in discrete
frequency bands (fixed or dynamically detennined or both fixed and dynamically
detennined bands) rather than the full bandwidth. This alternative approach would
take into account more than one audio steam in different frequency bands rather than
assuming that only a single stream is perceptible at a particular time.
Even a simple and computationally efficient process according to an aspect of
the present invention for segmenting audio has been found useful to identify auditoiy
events.
An auditoiy event detecting process of the present invention may be
implemented by dividing a time domain audio waveform into time intervals or blocks
and then converting the data in each block to the frequency domain, using either a
filter bank or a time-frequency transformation, such as a Discrete Fourier Transform
(DFT) (implemented as a Fast Fourier Transform (FFT) for speed). The amplitude of
the spectral content of each block may be normalized in order to eliminate or reduce
the effect of amplitude changes. The resulting frequency domain representation
provides an indication of the spectral content (amplitude as a function of frequency)

of the audio in the particular block. The spectral content of successive blocks is
compared and a change greater than a threshold may be taken to indicate the
temporal start or temporal end of an auditory event.
In order to minimize the computational complexity, only a single band of
frequencies of the time domain audio waveform may be processed, preferably either
the entire frequency band of the spectrum (which may be about 50 Hz to 15 kHz in
the case of an average quality music system) or substantially the entire frequency
band (for example, a band defining filter may exclude the high and low frequency
extremes).
Preferably, the frequency domain data is normalized, as is described below.
The degree to which the frequency domain data needs to be normalized gives an
indication of amplitude. Hence, if a change in this degree exceeds a predetermined
threshold, that too may be taken to indicate an event boundary. Event start and end
points resulting from spectral changes and from amplitude changes may be ORed
together so mat event boundaries resulting from both types of change are identified.
In practical embodiments in which the audio is represented by samples divided
into blocks, each auditoiy event temporal start and stop point boundary necessarily
coincides with a boundary of the block into which the time domain audio waveform
is divided. There is a trade off between real-tune processing requirements (as larger
blocks require less processing overhead) and resolution of event location (smaller
blocks provide more detailed information on the location of auditory events).
As a further option, as suggested above, but at the expense of greater
computational complexity, instead of processing the spectral content of the time
domain waveform in a single band of frequencies, the spectrum of the time domain
waveform prior to frequency domain conversion may be divided into two or more
frequency bands. Each of the frequency bands may then be converted to the
frequency domain and processed as though it were an independent channel. The
resulting event boundaries may then be ORed together to define the event boundaries
for that channel. The multiple frequency bands may be fixed, adaptive, or a
combination of fixed and adaptive. Tracking filter techniques employed in audio

noise reduction and other arts, for example, may be employed to define adaptive
frequency bands (e.g., dominant simultaneous sine waves at 800 Hz and 2 kHz could
result in two adaptively-determined bands centered on those two frequencies).
Other techniques for providing auditory scene analysis may be employed to
identify auditory events in the present invention.


DESCRIPTION OF THE ACCOMPANYING DPRAWINGS
FIG. 1A is a flow chart showing the extraction of a signature from an audio
signal in accordance with the present invention. The audio signal may, for example,
represent music (e.g., a musical composition or "song").
FIG. 1B is a flow chart illustrating the correlation of two signatures in
accordance with the present invention.
FIG. 2 is a flow chart showing the extraction of audio event locations and the
optional extraction of dominant subbands from an audio signal in accordance with
the present invention.
FIG. 3 is a conceptual schematic representation depicting the step of spectral
analysis in accordance with the present invention.
FIGS. 4 A and 4B are idealized audio waveforms showing a plurality of audio
event locations or event borders in accordance with the present invention.
FIG. 5 is a flow chart showing in more detail the correlation of two signatures
in accordance with the correlation 4 of FIG. 2 of the present invention.
FIGS. 6A-D are conceptual schematic representations of signals illustrating
examples of signature alignment in accordance with the present invention. The
figures are not to scale. In the case of a digital audio signal represented by samples,
the horizontal axis denotes the sequential order of discrete data stored in each
signature array.
BEST MODE FOR CARRYING OUT THE INVENTION
In a practical embodiment of the invention, the audio signal is represented by
samples that are processed in blocks of 512 samples, which corresponds to about

11.6 msec of input audio at a sampling rate of 44.1 kHz. A block length having a
time less than the duration of the shortest perceivable auditoiy event (about 20 msec)
is desirable. It will be understood that the aspects of the invention are not limited to
such a practical embodiment. The principles of the invention do not require
arranging the audio into sample blocks prior to detennining auditoiy events, nor, if
they are, of providing blocks of constant length. However, to minimize complexity,
a fixed block length of 512 samples (or some other power of two number of samples)
is useful for three primary reasons. First, it provides low enough latency to be
acceptable for real-time processing applications. Second, it is a power-of-two
number of samples, which is useful for fast Fourier transform (FFT) analysis. Third,
it provides a suitably large window size to perform useful auditoiy scene analysis. '
In the following discussions, the input signals are assumed to be data with
amplitude values in the range [-1,+1].
Auditoiy Scene Analysis 2 (FIG. I A)
Following audio input data blocking (not shown), the input audio signal is
divided into auditoiy events, each of which tends to be perceived as separate, in
process 2 ("Auditoiy Scene Analysis") of FIG. 1A. Auditoiy scene analysis may be
accomplished by an auditoiy scene analysis (ASA) process discussed above.
Although one suitable process for perfonning auditoiy scene analysis is described in
further detail below, the invention contemplates that other useful techniques for
perfonning ASA may be employed.
FIG. 2 outlines a process in accordance with, techniques of the present
invention that may be used as the auditoiy scene analysis process of FIG. 1 A. The
ASA step or process 2 is composed of three general processing substeps. The first
substep 2-1 ("Perform Spectral Analysis") takes the audio signal, divides it into
blocks and calculates a spectral profile or spectral content for each of the blocks.
Spectral analysis transforms the audio signal into the short-term frequency domain.
This can be performed using any filterbank; either based on transforms or banks of
band-pass filters, and in either linear or warped frequency space (such as the Bark
scale or critical band, which better approximate the characteristics of the human ear).

With any filterbank there exists a tradeoff between time and frequency. Greater time
resolution, and hence shorter time intervals, leads to lower frequency resolution.
Greater frequency resolution, and hence narrower subbands, leads to longer time
intervals.
The first substep 2-1 calculates the spectral content of successive time
segments of the audio signal. In a practical embodiment, described below, the ASA
block size is 512 samples of the input audio signal (FIG.3). In the second substep 2-
2, the differences in spectral content from block to block are determined ("Perform
spectral profile difference measurements"). Thus, the second substep calculates the
difference in spectral content between successive tune segments of the audio signal.
In the third substep 2-3 ("Identify location of auditory event boundaries"), when the
spectral difference between one spectral-profile block and the next is greater than a
threshold, the block boundary is taken to be an auditoiy event boundary. Thus, the
third substep sets an auditory event boundary between successive time segments
when the difference in the spectral profile content between such successive time
segments exceeds a threshold. As discussed above, a powerful indicator of the
beginning or end of a perceived auditoiy event is believed to be a change in spectral
content. The locations of event boundaries are stored as a signature. An optional
process step 2-4 ("Identify dominant subband") uses the spectral analysis to identify
a dominant frequency subband that may also be stored as part of me signature.
In this embodiment, auditoiy event boundaries define auditoiy events having a
length that is an integral multiple of spectral profile blocks with a minimum length of
one spectral profile block (512 samples in this example). In principle, event
boundaries need not be so limited.
Either overlapping or non-overlapping segments of the audio may be
windowed and used to compute spectral profiles of the input audio. Overlap results
in finer resolution as to the location of auditory events and, also, makes it less likely
to miss an event, such as a transient. However, as time resolution increases,
frequency resolution decreases. Overlap also increases computational complexity.
Thus, overlap may be omitted. FIG. 3 shows a conceptual representation of non-

overlapping 512 sample blocks being windowed and transformed into the frequency
domain by the Discrete Fourier Transform (DFT). Each block may be windowed and
transformed into the frequency domain, such as by using the DFT, preferably
implemented as a Fast Fourier Transform (FFT) for speed.
The following variables may be used to compute the spectral profile of the
input block:
N = number of samples in the input signal
M = number of windowed samples used to compute spectral profile
P = number of samples of spectral computation overlap
Q = number of spectral windows/regions computed
In general, any integer numbers may be used for the variables above.
However, the implementation will be more efficient if M is set equal to a power of 2
so that standard FFTs may be used for the spectral profile calculations. In a practical
embodiment of the auditory scene analysis process, the parameters listed may be set
to:
M =512 samples (or 11.6 msec at 44.1 kHz)
P =0 samples (no overlap)
The above-listed values were determined experimentally and were found
generally to identify with sufficient accuracy the location and duration of auditory
events. However, setting the value of P to 256 samples (50% overlap) has been
found to be useful in identifying some hard-to-find events. While many different
types of windows may be used to minimize spectral artifacts due to windowing, the
window used in the spectral profile calculations is an M-point Harming, Kaiser-
Bessel or other suitable, preferably non-rectangular, window. The above-indicated
values and a Hanning window type were selected after extensive experimental
analysis as they have shown to provide excellent results across a wide range of audio
material. Non-rectangular windowing is preferred for the processing of audio signals
with predominantly low frequency content. Rectangular windowing produces
spectral artifacts that may cause incorrect detection of events. Unlike certain codec
applications where an overall overlap/add process must provide a constant level, such

a constraint does not apply here and the window may be chosen for characteristics
such as its time/frequency resolution and stop-band rejection.
In substep 2-1 (FIG, 2), the spectrum of each M-sample block may be
computed by windowing the data by an M-point Harming, Kaiser-Bessel or other
suitable window, converting to the frequency domain using an M-point Fast Fourier
Transform, and calculating the magnitude of the FFT coefficients. The resultant data
is normalized so that the largest magnitude is set to unity, and the normalized array
of M numbers is converted to the log domain. The array need not be converted to the
log domain, but the conversion simplifies the calculation of the difference measure in
substep 2-2. Furthermore the log domain more closely matches the log domain
amplitude nature of the human auditory system. The resulting log domain values
have a range of minus infinity to zero. In a practical embodiment, a lower limit can
be imposed on the range of values; the limit may be fixed, for example -60 dB, or be
frequency-dependent to reflect the lower audibility of quiet sounds at low and very
high frequencies. (Note that it would be possible to reduce the size of the array to
M/2 in that the FFT represents negative as well as positive frequencies).
Substep 2-2 calculates a measure of the difference between the spectra of
adjacent blocks. For each block, each of the M (log) spectral coefficients from
substep 2-1 is subtracted from the corresponding coefficient for the preceding block,
and the magnitude of the difference calculated (the sign is ignored). These M
differences are then summed to one number. Hence, for the whole audio signal, the
result is an array of Q positive numbers; the greater the number the more a block
differs in spectrum from the preceding block. This difference measure could also be
expressed as an average difference per spectral coefficient by dividing the difference
measure by the number of spectral coefficients used in the sum (in this case M
coefficients).
Substep 2-3 identifies the locations of auditoiy event boundaries by applying a
threshold to the array of difference measures from substep 2-2 with a threshold value.
When a difference measure exceeds a threshold, the change in spectrum is deemed
sufficient to signal a new event and the block number of the change is recorded as an

event boundary. For the values of M and P given above and for log domain values
(in substep 2-1) expressed in units of dB, the threshold may be set equal to 2500 if
the whole magnitude FFT (including the mirrored part) is compared or 1250 if half
the FFT is compared (as noted above, the FFT represents negative as well as positive
frequencies — for the magnitude of the FFT, one is the mirror image of the other).
This value was chosen experimentally and it provides good auditory event boundary
detection. This parameter value may be changed to reduce (increase the threshold) or
increase (decrease the threshold) the detection of events. The details of this practical
embodiment are not critical. Other ways to calculate the spectral content of
successive time segments of the audio signal, calculate the differences between
successive time segments, and set auditory event boundaries at the respective
boundaries between successive time segments when the difference in the spectral
profile content between such successive time segments exceeds a threshold may be
employed.
For an audio signal consisting of Q blocks (of size M samples), the output of
the auditory scene analysis process of function 2 of FIG. 1A is an array B(q) of
information representing the location of auditoiy event boundaries where q = 0, 1,. .
. , Q-1. For a block size of M = 512 samples, overlap of P = 0 samples and a signal-
sampling rate of 44.1kHz, the auditory scene analysis function 2 outputs
approximately 86 values a second. Preferably, the array B{q) is stored as the
signature, such that, in its basic form, without the optional dominant subband
frequency information, the audio signal's signature is an array B(q) representing a
string of auditoiy event boundaries.
An example of the results of auditoiy scene analysis for two different signals is
shown in FIGS. 4A and 4B. The top plot, FIG. 4A, shows the results of auditoiy
scene processing where auditoiy event boundaries have been identified at samples
1024 and 1536. The bottom plot, FIG. 4B, shows the identification of event
boundaries at samples 1024, 2048 and 3072.

Identify dominant subband (optional)
For each block, an optional additional step in the ASA processing (shown in
FIG. 2) is to extract information from the audio signal denoting the dominant
frequency "subband" of the block (conversion of the data in each block to the
frequency domain results in information divided into frequency subbands). This
block-based information may be converted to auditory-event based information, so
that the dominant frequency subband is identified for every auditory event. This
information for every auditory event provides the correlation processing (described
below) with further information in addition to the auditory event boundary
information.
The dominant (largest amplitude) subband may be chosen from a plurality of
subbands, three or four, for example, that are within the range or band of frequencies
where the human ear is most sensitive. Alternatively, other criteria may be used to
select the subbands. The spectrum may be divided, for example, into three subbands.
The preferred frequency range of the subbands is:
Subband 1 301Hz to 560Hz
Subband 2 560Hz to 1938Hz
Subband 3 193 SHz to 9948Hz
To determine the dominant subband, the square of the magnitude spectrum (or
the power magnitude spectrum) is summed for each subband. This resulting sum for
each subband is calculated and the largest is chosen. The subbands may also be
weighted prior to selecting the largest. The weighting may take the form of dividing
the sum for each subband by the number of spectral values in the subband, or
alternatively may take the fonn of an addition or multiplication to emphasize the
importance of a band over another. This can be useful where some subbands have
more energy on average than other subbands but are less perceptually important.
Considering an audio signal consisting of Q blocks, the output of the dominant
subband processing is an array DS(q) of information representing the dominant
subband in each block (g = 0, 1,. . ., Q-1). Preferably, the array DS(q) is stored in
the signature along with the array B(q). Thus, with the optional dominant subband

information, the audio signal's signature is two arrays B(q) and DS(q), representing,
respectively, a string of auditory event boundaries and a dominant frequency subband
within each block. Thus, in an idealized example, the two arrays could have the
following values (for a case in which there are three possible dominant subbands).

In most cases, the dominant subband remains the same within each auditoiy
event, as shown in this example, or has an average value if it is not uniform for all
blocks within the event. Thus, a dominant subband may be determined for each
auditory event and the array DS(q) may be modified to provide that the same
dominant subband is assigned to each block within an event.
Correlation
The detennination of whether one signature is the same or similar to another
stored signature may be accomplished by a correlation function or process. The
correlation function or process compares two signatures to determine their similarity.
This may be done in two steps as shown in FIG. 5: a step 5-1 that removes or
minimizes the effect of temporal shift or delay on the signatures, followed by a step
5-2 that calculates a measure of similarity between the signatures.
The first-mentioned step 5-1 minimizes the effect of any delay between two
signatures. Such delay may have been deliberately added to the audio signal or could
be the result of signal processing and/or low bit rate audio coding. The output of this
step is two modified signatures in a form suitable for calculation of a measure of their
similarity.
The second-mentioned step 5-2 compares the two modified signatures to find a
quantative measure of their similarity (a correlation score). This measure of
similarity can then be compared against a threshold to determine if the signatures are
the same or different to a desired level of confidence. Two suitable correlation
processes or functions are described. Either one of them or some other suitable
correlation process or function may be employed as part of the present invention.

First Con-elation Process or Function
Removal of Temporal Delay Effects
This con-elation function or process isolates a single region or portion from
each of the signatures such that these two regions are the most similar portions in the
respective signatures and have the same length. The isolated region could be the
total overlapping regions between the two signatures, as shown in the examples in
FIGS. 6A-D, or the isolated region could be smaller than the overlapping regions.
The preferred method uses the whole overlapping region from the two
signatures. Some examples are shown in FIG. 6. The overlapping region for the two
signatures could be a portion from the end of one signature and the begirming of the
other signature (FIGS. 6B and 6C). If one of the signatures is smaller that the other,
then the overlapping region between the two signatures could be all of the smaller
signature and a portion of the larger signature (FIG. 6A and 6D).
There are a number of different ways to isolate a common region from two
arrays of data. A standard mathematical technique involves using the cross
con-elation to. find a lag or delay measure between the arrays of data. When the
beginning of each of two arrays of data is aligned, the lag or delay is said to be zero.
When the beginning of each of two arrays of data is not aligned, the lag or delay is
non-zero. The cross con-elation calculates a measure for each possible lag or delay
between the two arrays of data: this measure is stored as an array (the output of the
cross correlation function). The lag or delay that represents the peak in the cross
correlation array is considered to be the lag or delay of one array of data with respect
to the other. The following paragraphs expresses such a correlation method in a
mathematical form.
Let S1 (length N1) be an array from Signature 1 and S2 (length N2) an array
from Signature 2. First calculate the cross-correlation array RE1ε (see, for example,
John G. Proakis, Dimitris G. Manolakis, Digital Signal Processing: Principles,
Algorithms, and Applications, Macmillan Publishing Company, 1992, ISBN 0-02-
396815-X).


Preferably, the cross-correlation is performed using standard FFT based
techniques to reduce execution time.
Since both S1 and S2 are bounded, RE1E1 has length N1+N2-1. Assuming S1
and S2 are similar, the lag / corresponding to the maximum element in RE1E1
represents the delay of S2 relative to S1.

Since this lag represents the delay, the common spatial regions or spatially
overlapping parts of signatures S1 and S2 are retained as S'1 and S'2; each having the
same length, N12.
Expressed as equations, the overlapping parts S'1 and S'2 of the signatures S1
and S2 are defined as:

First Correlation Process or Function
Similarity Measure
This step compares the two signatures to find a quantitative measure of their
similarity. The preferred method uses the coefficient of correlation (Eqn. 5). This is

a standard textbook method (William Mendenhall, Dennis D. Wackerly, Richard L.
Scheaffer, Mathematical Statistics with Applications: Fourth Edition, Duxbury Press,
1990, ISBN 0-534-92026-8).

In practice, the value of the threshold may be tuned (on a large training set of
signatures) to ensure acceptable false rejection and detection rates.
The first correlation process or function is preferred for signatures that have
large misalignment or delay, and for signatures in which the length of one signature
is significantly smaller than the length of the other signature.

Second Correlation Process or Function
Removal of Temporal Delay Effects
The second correlation process or function transforms the signatures from their
current temporal domain into a domain that is independent of temporal delay effects.
The method results in two modified signatures that have the same length, such that
they can be directly correlated or compared.
There are a number of ways to transform data in such a manner. The preferred
method uses the Discrete Fourier Transform (DFT). The DFT of a signal can be
separated into magnitude and phase. A spatial shift or time delay of the signal (input
to the DFT) alters the phase of the DFT but not the magnitude. Thus the magnitude
of the DFT of a signal can be considered as a time-invariant representation of the
signal.
This property of the DFT allows each of the two signatures to be transformed
into a time-invariant representation. If both signatures have the same length, the
magnitude DFT can be directly computed for each of the signatures and the results
stored as the modified signatures. If the length of each of the signatures is different,
then prior to calculating the DFT, either the longer signature can be truncated to have
the same length as the shorter signature, or the shorter signature can be zero padded
or extended to have the same length as the longer signature. The following
paragraphs express the method in a mathematical form.
Let S1 (length TV,) be an array from Signature 1 and S2 (length N2) an array
from Signature 2. Firstly, the longer signature is truncated or the shorter signature
zero padded such that both signatures have the same length, N12. The transformed
signature arrays, S'1 and S'2, are created by taking the magnitude DFT as follows:

In practice, for each signature it is beneficial to subtract its mean prior to
calculating the DFT. Some windowing may also be applied to the S1 and S2

signatures prior to taking the Discrete Fourier Transform, however in practice no
particular windowing has been found to produce the best results.
Second Correlation Process or Function
Similarity Measure
This similarity measure step compares the two signatures to find a quantitative
measure of their similarity. The preferred method uses the coefficient of correlation
(Eqn. 9). This is a standard textbook method (William Mendenhall, Dennis D.
Wackerly, Richard L. Scheaffer, Mathematical Statistics with Applications: Fourth
Edition, Duxbury Press, 1990, ISBN 0-534-92026-8).

In practice, the value of the threshold may be tuned (on a large training set of
signatures) to ensure acceptable false rejection and detection rates.

In practical applications, many signatures may be stored together to form a
library of signatures representing "known" audio content. In this situation, the ability
to discriminate between signatures can be improved by calculating a mean signature
and subtracting this mean signature from each of two signatures under comparison.
For example, given a database containing W signatures, S'0 to S'w-1 the mean
signature is calculated as follows.

When comparing two signatures (even if one of the signatures is not in the
library) the mean signature is subtracted from both signatures prior to calculating the
covariance (subsequently used in the coefficient of correlation). The covariance
becomes:

The second correlation process or function is preferred for signatures that have
small misalignment or delay, and for signatures where the lengths of the signatures
are similar. It is also significantly faster than the first correlation process or function.
However since some information is inherently lost (by discarding the phase of the
DFTs), it results in a slightly less accurate measure of similarity.
Applications
As briefly mentioned earlier, an application of this invention is searchable
audio databases; for example a record company's library of songs. Signatures could
be created for all the songs from the library and the signatures stored in a database.
This invention provides a means for taking a song of unknown origin, calculating its

signature and comparing its signature very quickly against all the signatures in the
database to determine the identity of the unknown song.
In practice, the accuracy of (or confidence in) the similarity measure is
proportional to the size of the signatures being compared. The greater the length of
the signatures, the greater the amount of data being used in the comparison and hence
the greater the confidence or accuracy in the similarity measure. It has been found
that signatures generated from about 30 seconds of audio provide for good
discrimination. However the larger the signatures, the longer the time required to
perform a comparison.
Conclusion
It should be understood that implementation of other variations and
modifications of the invention and its various aspects will be apparent to those skilled
in the art, and that the invention is not limited by these specific embodiments
described. It is therefore contemplated to cover by the present invention any and all
modifications, variations, or equivalents that fall within the true spirit and scope of
the basic underlying principles disclosed and claimed herein.
The present invention and its various aspects may be implemented as software
functions performed in digital signal processors, programmed general-purpose digital
computers, and/or special purpose digital computers. Interfaces between analog and
digital signal streams may be performed in appropriate hardware and/or as functions
in software and/or firmware.

WE CLAIM:
1. A method for determining if one audio signal is derived from another
audio signal or if two audio signals are derived from the same audio signal,
comprising
comparing reduced-information characterizations of said audio signals,
wherein said reduced-information characterizations represent at least the
boundaries of auditory events resulting from the division of each of said audio
signals into auditory events, each of which auditory events tends to be perceived as
separate and distinct, wherein each audio signal is divided into auditory events by
detecting changes in signal characteristics with respect to time in the audio
signal, and
identifying a continuous succession of auditory event boundaries in the
audio signal, in which every change in signal characteristics with respect to time
exceeding a threshold defines a boundary, wherein each auditory event is an audio
segment between adjacent boundaries and there is only one auditory event between
such adjacent boundaries, each boundary representing the end of the preceding
event and the beginning of the next event such that a continuous succession of
auditory events is obtained, wherein neither auditory event boundaries, auditory
events, nor any characteristics of an auditory event are known in advance of
identifying the continuous succession of auditory event boundaries and obtaining
the continuous succession of auditory events.
2. The method as claimed in claim 1, wherein said comparing is carried out
by removing from the characterizations or minimizing in the characterizations the
effect of temporal shift or delay an the audio signals, calculating a measure of
similarity, and comparing the measure of similarity against a threshold.

3. The method as claimed in claim 2 wherein said removing identifies a
portion in each of said characterizations, such that the respective portions are the
most similar portions in the respective characterizations and the respective portions
have the same length.
4. The method as claimed in claim 3, wherein said removing identifies a
portion in each of said characterizations by performing a cross-correlation.
5. The method as claimed in claim 4, wherein said calculating calculates a
measure of similarity by calculating a coefficient of correlation of the identified
portion in each of said characterizations.
6. The method as claimed in claim 2, wherein said removing transforms die
characterizations into a domain that is independent of temporal delay effects.
7. The method as claimed in claim 6, wherein said removing transforms the
characterizations into the frequency domain.
8. The method as claimed in claim 7, wherein said calculating calculates a
measure of similarity by calculating a coefficient of correlation of an identified
portion in each of said characterizations.
9. The method as claimed in claim 1, wherein one of said characterizations
is a characterization from a library of characterizations representing known audio
content.

10. The method as claimed in claim 9, which involves subtracting a mean of
the characterizations in said library from bath characterizations after said removing
and prior to said comparing.
11. The method as claimed in claim 1, wherein said reduced-information
characterizations also represent the dominant frequency subband of auditory
events.

A method for determining if one audio signal is derived from another audio
signal or if two audio signals are derived from the same audio signal compares
reduced-information characterizations of said audio signals, wherein said
characterizations are based on auditory scene analysis. The comparison removes
from the characterisations or minimizes in the characterisations the effect of
temporal shift or delay on the audio signals (5-1), calculates a measure of
similarity (5-2), and compares the measure of similarity against a threshold. In one
alternative, the effect of temporal shift or delay is removed or minimized by cross-
correlating the two characterizations. In another alternative, the effect of temporal
shift or delay is removed or minimized by transforming the characterizations into a
domain that is independent of temporal delay effects, such as the frequency
domain. In both cases, a measure of similarity is calculated by calculating a
coefficient of correlation.

Documents:

1488-kolnp-2003-abstract.pdf

1488-kolnp-2003-assignment.pdf

1488-kolnp-2003-assignment1.1.pdf

1488-kolnp-2003-claims.pdf

1488-KOLNP-2003-CORRESPONDENCE 1.1.pdf

1488-KOLNP-2003-CORRESPONDENCE 1.2.pdf

1488-kolnp-2003-correspondence.pdf

1488-kolnp-2003-correspondence1.3.pdf

1488-kolnp-2003-description (complete).pdf

1488-KOLNP-2003-DESCRIPTION(COMPLETE)1.1.pdf

1488-KOLNP-2003-DRAWING 1.1.pdf

1488-kolnp-2003-drawings.pdf

1488-kolnp-2003-examination report.pdf

1488-kolnp-2003-examination report1.1.pdf

1488-KOLNP-2003-FORM 1.1.pdf

1488-kolnp-2003-form 1.pdf

1488-kolnp-2003-form 18.1.pdf

1488-kolnp-2003-form 18.pdf

1488-kolnp-2003-form 3.1.pdf

1488-kolnp-2003-form 3.pdf

1488-kolnp-2003-form 5.1.pdf

1488-kolnp-2003-form 5.pdf

1488-KOLNP-2003-FORM-27.pdf

1488-kolnp-2003-gpa.pdf

1488-kolnp-2003-gpa1.1.pdf

1488-kolnp-2003-granted-abstract.pdf

1488-kolnp-2003-granted-claims.pdf

1488-kolnp-2003-granted-description (complete).pdf

1488-kolnp-2003-granted-drawings.pdf

1488-kolnp-2003-granted-form 1.pdf

1488-kolnp-2003-granted-specification.pdf

1488-KOLNP-2003-OTHERS 1.1.pdf

1488-kolnp-2003-others.pdf

1488-kolnp-2003-reply to examination report.pdf

1488-kolnp-2003-reply to examination report1.1.pdf

1488-kolnp-2003-specification.pdf


Patent Number 248786
Indian Patent Application Number 1488/KOLNP/2003
PG Journal Number 34/2011
Publication Date 26-Aug-2011
Grant Date 24-Aug-2011
Date of Filing 17-Nov-2003
Name of Patentee DOLBY LABORATORIES LICENSING CORPORATION
Applicant Address 100 POTRERO AVENUE, SAN FRANCISCO, CA
Inventors:
# Inventor's Name Inventor's Address
1 CROCKETT BRETT G 100 POTRERO AVENUE, SAN FRANCISCO, CA 94103
2 SMITHERS MICHAEL J 100 POTRERO AVENUE, SAN FRANCISCO, CA 94103
PCT International Classification Number G10L 11/00
PCT International Application Number PCT/US2002/05329
PCT International Filing date 2002-02-22
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 60/293,825 2001-05-25 U.S.A.
2 10/045,644 2002-01-11 U.S.A.
3 60/351,498 2002-01-23 U.S.A.
4 PCT/US02/04317 2002-02-12 U.S.A.