EP1388145B1 - Device and method for analysing an audio signal in view of obtaining rhythm information - Google Patents

Abstract

An apparatus for analyzing an audio signal with regard to rhythm information of the audio signal comprises a filterbank for dividing the audio signal into at least two sub-band signals. Every sub-band signal is examined with regard to a periodicity of the sub-band signal to obtain rhythm raw-information of every sub-band signal. The rhythm raw-information is subjected to a quality evaluation to obtain a significance measure for every sub-band signal. The rhythm information of the audio signal will finally be established by considering the significance measure of the sub-band signal and the rhythm raw-information. This enables a more robust analysis of the audio signal, since sub-band signals, where significant rhythm information are present, are preferred compared to sub-band signals where less significant rhythm information are present, when establishing the rhythm information.

Description

The present invention relates to signal processing concepts and especially on the analysis of audio signals regarding rhythm information.

In recent years, the availability of multimedia data, such as B. audio or video data, has risen sharply. This is due to a number of technical factors which particularly affects the wide availability of the Internet, of powerful computer hardware and Software and powerful data compression processes, d. H. Source coding of audio and video processes support.

The huge amounts of audiovisual data, for example available on the Internet worldwide, require concepts which make this data possible according to content criteria assess, catalog, etc. It exists the desire to be able to target multimedia data to search and find by specifying meaningful criteria.

This requires so-called "content-based" techniques that from the audiovisual data so-called characteristics that are in professional circles also called "features", extract, what important characteristic properties of the signal represent. Based on such features or combinations These characteristics can have similarity relationships or similarities derived between audio or video signals become. This is done by comparing or relating the extracted feature values from the different Signals, which are simply referred to as "pieces" become.

The determination or extraction is of particular interest of characteristics that are not only signal-theoretical, but as possible have immediate semantic meaning, d. H. from Represent listeners' properties that are immediately felt.

This allows the user to be simple and intuitive Way to formulate search queries to pieces from across the to find existing data from an audio signal database. Semantically relevant features, similarity relationships also allow it to model between pieces that the come close to human sensation. The use of features which have semantic meaning also makes it possible, for example an automatic suggestion for a specific one Users interesting pieces if his preferences are known.

In the field of music analysis, the tempo is an important musical one Parameter that has semantic meaning. The tempo is usually measured in "beats per minute" (BPM). The automatic extraction of the tempo as well as the focal points the "beat" or generally speaking the automatic extraction of rhythm information, is an example of the extraction of a semantically important feature of a piece of music.

There is also a desire that feature extraction, d. H. extracting rhythm information an audio signal, robust and computationally efficient can. Robustness means that it doesn't matter whether the piece has been source coded and decoded again, whether the piece played through a loudspeaker and from a Mi microphone received whether it was played loudly or softly or whether it is from one instrument or 'a plurality of Instruments is played.

For the determination of the center of gravity and thus also the Tempo, d. H. for the determination of rhythm information, the term "beat tracking" has also established. It is already known from the prior art Beat tracking based on a note-like or transcribed Signal representation, e.g. B. in midi format. However, the goal is to avoid such meta representations need, but an analysis directly with a z. B. PCM encoded or generally speaking digitally present Audio signal.

The specialist publication "Tempo and Beat Analysis of Acoustic Musical Signals "by Eric D. Scheirer, J. Acoust. Soc. Am. 103: 1, (Jan 1998), pages 588-601, discloses a method for the automatic extraction of a rhythmic pulse from musical Excerpts. The input signal is generated using a Filter bank split into a number of sub-bands, for example in 6 subbands with crossover frequencies of 200 Hz, 400 Hz, 800 Hz, 1600 Hz and 3200 Hz. For the first subband low pass filtering is performed. For the last one Sub-band high pass filtering is performed for which The remaining sub-volumes in between are bandpass filtered described. Each sub-band is processed as follows. The subband signal is first rectified. In in other words, the absolute value of the samples certainly. The resulting n values are then smoothed, for example with an averaging over a suitable one Window to get an envelope signal. To lower the Computational complexity can undersampled the envelope signal become. The envelope signals are differentiated, i. H. sudden changes in signal amplitude are caused by the Differentiation filter preferably forwarded. The result is then limited to non-negative values. Any envelope signal is then placed in a bank of resonant filters, i.e. H. oscillators, given a filter for each tempo range included so that it fits the musical tempo Filter is most excited. For each filter, the Energy of the output signal as a measure of the match the tempo of the input signal with that belonging to the filter Tempo calculated. The energies for each pace will eventually be summed up over all subbands, the largest energy sum the pace delivered as a result, d. H. the rhythm information, features.

A major disadvantage of this method is that great computing and storage complexity, especially for implementation the large number of parallel oscillating "oscillators", ultimately only one is selected. This makes an efficient implementation for example for Real-time applications almost impossible.

The specialist publication "Pulse Tracking with a Pitch Tracker" by Eric D. Scheirer, Proc. 1997 Workshop on Applications of Signal Processing to Audio and Acoustics, Mohonk, NY, Oct. 1997, describes a comparison of the "oscillator concept" described above with an alternative concept based on the Use of autocorrelation functions to extract the Periodicity from an audio signal, d. H. the rhythm information of a signal. An algorithm for modulation human pitch perception, d. H. of the pitch used for "beat tracking".

The known algorithm is shown in FIG. 3 as a block diagram. The audio signal becomes one via an audio input 300 Analysis filter bank 302 fed. The analysis filter bank generates a number n of channels from the audio input, i. H. of individual subband signals. Each subband signal contains a certain range of frequencies of the audio signal. The Filters of the analysis filter bank are selected so that they Approach the selection characteristics of the human inner ear. Such an analysis filter bank is also called a gamma-tone filter bank designated.

In the devices 304a to 304c the rhythm information evaluated each subband signal. For each The input signal first becomes an envelope-like output signal calculated (according to a so-called "Inner Hair Cell "processing in the ear) and subsampled. From this result an autocorrelation function (AKF) is calculated to the periodicity of the signal as a function of delay, d. H. to get the "lag".

The output of the devices 304a to 304c then lies for each Subband signal an autocorrelation function, which Represents aspects of the rhythm information of each subband signal.

The individual autocorrelation functions of the subband signals are then combined in a device 306 by summation, to get a sum auto-correlation function (SAKF), which is the rhythm information of the signal at the audio input 300 reproduces. This information can be found on a Tempo output 308 are output. Large values in the sum autocorrelation indicate that for a one tip the SAKF associated delay (lag) a high periodicity of Beginnings of grades are available. Therefore, for example, the largest Value of the sum auto-correlation function within the musical reasonable delays sought.

Musically sensible delays are, for example Tempo range between 60 bpm and 200 bpm. The device 306 can also be arranged to include a delay time in tempo information implement. For example, one Peak a delay of one second at a tempo of 60 Beats per minute. Smaller delays indicate higher ones Tempos down, while major delays to smaller tempos than 60 bpm.

This method has compared to the first method an advantage in that no large oscillators Computing and storage effort must be implemented. on the other hand the concept is disadvantageous in that the Quality of the results depends very much on the type of audio signal depends. For example, is a dominant from an audio signal To hear the rhythm instrument, this is shown in Fig. 3 described concept work well. Is against that Voice dominant, which has no particularly clear rhythm information will deliver, so the rhythm determination be ambiguous. In the audio signal could also be a Band that only contains rhythm information, such as B. a higher frequency band in which, for example, a Hihat a drum kit is positioned, or a low one Frequency band in which the bass drum of a drum set the frequency scale is positioned. Because of the combination However, the individual information becomes the somewhat unambiguous Information of these special subbands from the ambiguous Information of the other subbands overlaid or "Watered down".

Another problem with using autocorrelation functions to extract the periodicity of a subband signal is that the sum autocorrelation function, obtained by means 306 is ambiguous. The sum autocorrelation function at output 306 is to that effect ambiguous that even when multiplying a delay an auto-correlation function spike is generated. This is understandable that a sine component with a period from t0 when processing autocorrelation function is subjected, in addition to the desired maximum at t0 also maxima at multiples of the delays, i.e. H. at 2t0, 3t0, etc. generated.

The specialist publication "A Computationally Efficient Multipitch Analysis Model ", by Tolonen and Karjalainen, IEEE Transactions on Speech and Audio Processing, Volume 8, No. 6, Nov. 2000, discloses a computationally efficient model for a periodicity analysis of complex audio signals. The computing model divides the signal into two channels, one Channel below 1000 Hz and one channel above 1000 Hz. This becomes an autocorrelation of the lower channel and an autocorrelation the envelope of the upper channel. Finally the two autocorrelation functions are summed up. To the Eliminate ambiguities in the sum autocorrelation function, the sum autocorrelation function is processed further, a so-called enhanced summary autocorrelation Function (ESACF) (further developed sum autocorrelation function) to obtain. This postprocessing of the sum autocorrelation function involves repeated subtraction of versions of the autocorrelation function spread with integer factors from the sum auto-correlation function with subsequent limitation to non-negative values.

The object of the present invention is to make computing time efficient and robust device and a computationally efficient and robust method for analyzing an audio signal regarding rhythm information.

This task is accomplished by a device for analyzing a Audio signal according to claim 1 or by a method for analyzing an audio signal according to claim 11 solved.

The present invention is based on the knowledge that in the individual frequency bands, d. H. the sub-bands, often different favorable conditions for finding of rhythmic periodicities. While for example with pop music often in the middle, for example around 1 kHz, the signal from not corresponding to the beat Singing is dominated are in the higher frequency ranges often especially percussion sounds present, such as B. the hihat of the drums, which is a very good extraction allow rhythmic regularities. In other words, include different frequency bands depending on the audio signal a different amount of rhythmic information or have different quality or significance for that Rhythm information of the audio signal.

According to the invention, the audio signal is therefore first divided into subband signals disassembled. Each subband signal is regarding its periodicity examined to provide raw rhythm information to get for each subband signal. Then according to the present invention an assessment of the quality of periodicity each subband signal performed to a level of significance to get for each subband signal. A high level of significance indicates that in this subband signal there is clear rhythm information, while a low Significance level suggests that in this subband signal less clear rhythm information is available.

According to a preferred embodiment of the present Invention is being used in the investigation of a subband signal of its periodicities a modified one Envelope of the subband signal is calculated and then an autocorrelation function the envelope is calculated. The autocorrelation function the envelope provides the raw rhythm information There is clear rhythm information when the autocorrelation function has clear maxima while less there is clear rhythm information when the autocorrelation function the envelope of the subband signal less pronounced signal peaks or no signal peaks at all Has. An autocorrelation function that clearly peaks has a high degree of significance, while an autocorrelation function that has a relatively flat History, a low level of significance is obtained.

According to the invention, the individual rhythm raw information of the individual subband signals not simply "blindly" combined, but taking into account the measure of significance for each sub-band signal used the rhythm information to get the audio signal. A subband signal is high Significance measure, so it will be when determining the rhythm information preferred while a subband signal is a has a low level of significance, d. H. that a low quality in terms of rhythm information when determining the rhythm information of the audio signal hardly or in extreme cases it is not considered at all.

This can be done in a time-efficient manner using a weighting factor be implemented, which depends on the significance measure. While a subband signal that is good quality for that Has rhythm information, d. H. which is a high level of significance has a weighting factor of 1, a another subband signal that has a smaller measure of significance, get a weighting factor less than 1. In extreme cases becomes a subband signal that has a completely flat autocorrelation function has a weighting factor of 0. The weighted auto-correlation functions, d. H. the weighted Rhythm raw information is then simply added up. If only one subband signal of all subband signals provides good rhythm information while the other subband signals Autocorrelation functions with a flat course in extreme cases, this weighting can lead to that all subband signals except the one subband signal get a weighting factor of 0, d. H. in the determination not taking the rhythm information into account at all so that the rhythm information of the audio signal only can be determined from a single subband signal.

The concept according to the invention is advantageous in that it enables a robust determination of the rhythm information, since subband signals with no clear or even deviating rhythm information, d. H. when singing one has a different rhythm than the actual beat of the Piece that does not "water down" the rhythm information of the audio signal or "falsify". They also become very intoxicating Subband signals, which have a system autocorrelation function deliver the signal / noise ratio with a completely flat curve in determining the rhythm information don't worsen. However, this is exactly what would occur if, as in the prior art, simply all auto-correlation functions of subband signals with the same weight be added up.

Another advantage of the method according to the invention is in that with a little extra computational effort Significance measure can be determined and that the assessment the rhythm raw information with the significance measure and the subsequent summation without much storage and computing time can be carried out efficiently, what the inventive Concept especially for real-time applications recommends.

Fig. 1

ein Blockschaltbild einer Vorrichtung zum Analysieren eines Audiosignals mit einer Qualitätsbewertung der Rhythmus-Rohinformationen;

Fig. 2

ein Blockschaltbild einer Vorrichtung zum Analysieren eines Audiosignals unter Verwendung von Gewichtungsfaktoren auf der Basis der Signifikanzmaße;

Fig. 3

ein Blockschaltbild einer bekannten Vorrichtung zum Analysieren eines Audiosignals hinsichtlich von Rhythmusinformationen;

Fig. 4

ein Blockschaltbild einer Vorrichtung zum Analysieren eines Audiosignals hinsichtlich von Rhythmusinformationen unter Verwendung einer Autokorrelationsfunktion mit einer teilbandweisen Nachbearbeitung der RhythmusRohinformationen; und

Fig. 5

ein detailliertes Blockschaltbild der Einrichtung zum Nachbearbeiten von Fig. 4.

Preferred embodiments of the present invention are explained in detail below with reference to the accompanying drawings. Show it:

Fig. 1

a block diagram of a device for analyzing an audio signal with a quality evaluation of the rhythm raw information;

Fig. 2

a block diagram of an apparatus for analyzing an audio signal using weighting factors based on the significance measures;

Fig. 3

a block diagram of a known device for analyzing an audio signal for rhythm information;

Fig. 4

a block diagram of a device for analyzing an audio signal for rhythm information using an autocorrelation function with a sub-band post-processing of the raw rhythm information; and

Fig. 5

4 shows a detailed block diagram of the device for reworking of FIG. 4.

1 shows a block diagram of a device for analysis an audio signal with respect to rhythm information. The audio signal is received via an input 100 of a device 102 for splitting the audio signal into at least two Subband signals 104a and 104b supplied. Any subband signal 104a, 104b is inserted into a device 106a or 106b 'for examination the same in terms of periodicities in the subband signal supplied to raw rhythm information 108a or 108b for each subband signal. The raw rhythm information are then a device 110a or 110b to assess a quality of the periodicity of each of the at least two subband signals fed to a significance measure 112a, 112b for each of the at least two subband signals. Both the raw rhythm information 108a, 108b and the significance measures 112a, 112b also become a device 114 for determining the rhythm information of the audio signal fed. The device 114 takes into account when determining the rhythm information of the audio signal the significance measures 112a, 112b for the subband signals and the raw rhythm information 108a, 108b of at least one subband signal.

For example, has the device 110a for quality evaluation found that in the subband signal 104a no particular Periodicity is present, so the significance measure 112a becomes very be small or equal to 0. In this case, the establishment 114 to determine the rhythm information that the significance measure 112a is zero, so that the rhythm raw information 108a of subband signal 104a during the determination the rhythm information of the audio signal at all no longer need to be considered. The rhythm information of the audio signal are then alone and exclusively based on raw rhythm information 108b of the subband signal 104b determined.

The following will refer to FIG. 2 for a special one Embodiment of the device of Fig. 1 received. As Means 102 for splitting the audio signal can be a conventional one Analysis filter bank used to be the output side a user selectable number of subband signals supplies. Each subband signal is then processed Devices 106a, 106b and 106c subjected, then by means 110a to 110c of any rhythm raw information Significance measures can be determined. The facility 114 includes the preferred one shown in FIG Embodiment means 114a for computing Weighting factors for each subband signal based on the Significance measure for this subband signal and optionally also of the other subband signals. In device 114b takes place then a weighting of the raw rhythm information 108a to 108c with the weighting factor for this subband signal instead, whereupon, also in the device 114b, the weighted raw rhythm information combined, e.g. B. summed up, to the rhythm information at the tempo output 116 to get the audio signal.

The concept according to the invention thus arises as follows After evaluating the rhythmic information of the Single tapes, which are generated, for example, by envelope formation, Smoothing, differentiation, limitation to positive values and Forming the autocorrelation function can take place (facilities 106a to 106c) finds an evaluation of the value or the quality of these interim results in the facilities 110a to 110c instead. This is done with the help of an evaluation function achieved which the reliability of each Individual results assessed with a significance measure. From the Significance measures of all subband signals become a weighting factor for each band for the extraction of the rhythm information derived. The overall result of rhythm extraction is then combined in the device 114b the band-by-band individual results, taking into account their respective weighting factors reached.

The result is an algorithm implemented in this way for rhythm analysis a good ability to get rhythmic information in a signal even under unfavorable conditions to find reliably. The concept according to the invention draws is therefore very robust.

In a preferred embodiment, the rhythm raw information 108a, 108b, 108c, which determine the periodicity of the represent the respective subband signal by means of an autocorrelation function certainly. In this case, it is preferred determine the significance measure by taking a maximum of the autocorrelation function by an average of the autocorrelation function is divided, and then the value 1 is subtracted becomes. It should be noted that any autocorrelation function always a local maximum with a delay of 0 that represents the energy of the signal. This maximum should be disregarded so that the quality determination is not falsified.

Furthermore, the autocorrelation function should only be in one special tempo range are considered, d. H. of a maximum Delay the slowest pace of interest corresponds to a minimum delay that corresponds to the highest pace of interest. A typical pace area is between 60 bpm and 200 bpm.

Alternatively, the ratio between the arithmetic mean of the autocorrelation function in the interest Tempo range and the geometric mean the autocorrelation function in the tempo area of interest be determined. It is known that when all values of the autocorrelation function are the same, d. H. if the autocorrelation function has a flat course, the geometric Average of the autocorrelation function and the arithmetic Mean of the autocorrelation function are the same. In this Case the significance measure would have a value equal to 1, which means that the rhythm raw information is not significant are.

In the case of a system auto-correlation function with strong The ratio of the arithmetic mean would peak geometric mean greater than 1, which means that the Autocorrelation function has good rhythm information. However, the smaller the ratio between the arithmetic mean and geometric mean, the flatter it is the autocorrelation function and the fewer periodicities contains them, which in turn means the rhythm information this subband signal is less significant, i. H. a have lower quality, which can be seen in a low or a Weighting factor of 0 will express.

There are various options with regard to the weighting factors. A relative weighting is preferred, such that all weighting factors of all subband signals add up to 1, d. H. that the weighting factor of a band is determined to be divided as the significance value of this band by the sum of all significance values. In this case is a relative weighting before the totalization of the weighted Rhythm raw information performed to the rhythm information to get the audio signal.

As has already been stated, it is preferred to evaluate the rhythm information using an autocorrelation function perform. This case is shown in Fig. 4 shown. The audio signal is via the audio signal input 100 in the device 102 for decomposing the audio signal fed into subband signals 104a and 104b. Any subband signal is then in the device 106a or 106b as it has been performed using an autocorrelation function examined the periodicity of the subband signal to investigate. At the exit of the device 106a or 106b the raw rhythm information 108a, 108b is then available. This are fed into a device 118a or 118b in order to by means of the facility's autocorrelation function 116a postprocessed raw rhythm information. So u. a. ensured that the ambiguities of the Autocorrelation function, d. H. that with integer multiples the delays also occur, can be eliminated on a sub-band basis to post-processed raw rhythm information To get 120a or 120b.

This has the advantage that the ambiguities of the autocorrelation functions, d. H. the raw rhythm information 108a, 108b, are already eliminated in sub-bands, and not only as in the prior art, after the summation of the individual autocorrelation functions. In addition, the single band allows Elimination of ambiguities in the autocorrelation functions by means 118a, 118b that the raw rhythm information of the subband signals independently can be handled by each other. For example, you can a quality assessment by means 110a for the rhythm raw information 108a or by means of the device 110b for the rhythm raw information 108b become.

As shown by the dashed lines in Fig. 4 However, the quality assessment can also be based on the postprocessed rhythm raw information take place, whereby This latter option is preferred because of the quality assessment on the basis of the reworked raw rhythm information ensures that the quality of information is judged that is no longer ambiguous.

The establishment of the rhythm information by the facility 114 then takes place on the basis of post-processed rhythm information of a channel and preferably also on the Basis of the significance measure for this channel instead.

If a quality assessment based on the raw rhythm information, that is, the signal before device 118a is carried out, it is advantageous in that if it is determined that the measure of significance is 0, i. H. that the autocorrelation function has a flat course has the post-processing by means of the device 118a can be completely dispensed with in order to save computing time resources.

In the following, Fig. 5 is discussed in order to provide a more detailed description Setup of a device 118a or 118b for post-processing the raw rhythm information. First the subband signal, for example 104a, into the device 106a for examining the periodicity of the subband signal fed by means of an autocorrelation function in order to To obtain rhythm raw information 108a. The ambiguity Eliminating part of a band can be the same as in the stand the technology, a spread autocorrelation function by means of a device 121 are calculated, the device 121 is arranged to the spread auto-correlation function to be calculated so that it is an integer multiple of one Delay is spread. A device 122 is in this Case arranged to use the spread auto-correlation function from the original autocorrelation function, d. H. the Subtract raw rhythm information 108a. In particular it is preferred to first double one To calculate autocorrelation function in the device 121 and then subtract from the raw rhythm information 108a. Then, in the next step, one becomes a factor of 3 spread autocorrelation function in the device 121 is calculated and from the result of the previous subtraction again subtracted so that gradually all ambiguities are eliminated from the raw rhythm information.

Alternatively or additionally, the device 121 can be arranged be an auto-correlation function compressed by an integer factor to calculate, this then from the facility 122 added to the raw rhythm information in order to also generate shares for delays t0 / 2, t0 / 3 etc.

In addition, the spread or compressed versions the rhythm raw information 108a before adding or Subtract to be weighted for flexibility to achieve in the sense of high robustness.

By the method, the periodicity of a subband signal to investigate on the basis of an autocorrelation function, a further improvement can be achieved if the Properties of the autocorrelation function are included and post-processing using the facility 118a or 118b is performed. So creates a periodic Sequence of note starts with a distance t0 not just one AKF peak with a delay t0 but also with 2t0, 3t0, etc. This becomes an ambiguity in tempo detection, d. H. the search for significant maxima in the autocorrelation function, to lead. This can eliminate the ambiguities when versions are spread by integer factors the AKF is subtracted from the initial value sub-band (weighted) become.

Another problem with the autocorrelation function is that that they have no information at t0 / 2, t0 / 3 ... etc., ie at Double, triple, etc. of the "basic tempo" delivers what is special can lead to wrong results if two Instruments that are in different sub-bands together define the rhythm of the signal. This thing will taking into account that upset factors Versions of the autocorrelation function are calculated and these are then weighted to the raw rhythm information or unweighted be added.

AKF post-processing therefore takes place sub-band, whereby an autocorrelation function for at least one subband signal is calculated and this with stretched or spread Versions of this feature is combined.

Claims (11)

1. Apparatus for analyzing an audio signal with regard to rhythm information of the audio signal, comprising:

   means (102) for dividing the audio signal into at least two sub-band signals (104a, 104b);

   means for examining (106a, 106b) a sub-band signal with regard to a periodicity in the sub-band signal, to obtain rhythm raw-information (108a, 108b) for the sub-band signal;

   means for evaluating (110a, 110b) a quality of the periodicity of the rhythm raw-information (108a) of the sub-band signal (104a) to obtain a significance measure (112a) for the sub-band signal; and

   means (114) for establishing rhythm information of the audio signal under consideration of the significance measure (112a) of the sub-band signal and the rhythm raw-information (108a, 108b) of at least one sub-band signal.
2. Apparatus according to claim 1, wherein the means for examining (106a, 106b) is formed to calculate an autocorrelation function for each of the least two sub-band signals.
3. Apparatus according to claim 1 or 2, wherein the means for examining (106a, 106b) comprises:

   means for forming an envelope of a sub-band signal;

   means for smoothing the envelope of the sub-band signal to obtain a smoothed envelope;

   means for differentiating the smoothed envelope to obtain a differentiated envelope;

   means for limiting the differentiated envelope to positive values to obtain a limited envelope; and

   means for forming an autocorrelation function of the limited envelope to obtain the rhythm raw-information (108a, 108b).
4. Apparatus according to claim 2 or 3, wherein the means for evaluating (110a, 110b) of the quality is formed to use a ratio of a maximum of the autocorrelation function to an average of the autocorrelation function as a significance measure.
5. Apparatus according to claim 2 or 3, wherein the means for evaluating (110a, 110b) of the quality is formed to use a ratio of an arithmetic average of the rhythm raw-information to a geometrical average of the rhythm raw-information as significance measure.
6. Apparatus according to claim 4 or 5, wherein the means for evaluating (110a, 110b) the quality is formed to evaluate the autocorrelation function merely within a tempo range, which extends from a minimum lag to obtain a maximum tempo to a maximum lag to obtain a minimum tempo.
7. Apparatus according to one of the previous claims, wherein means for establishing (114) comprises:

   means (114a) for deriving a weighting factor for a sub-band by using the significance measure for the sub-band;

   means (114b) for weighting a rhythm raw-information of the sub-band by using the weighting factor for the sub-band to obtain weighted rhythm raw-information for the sub-band and for summarizing the weighted rhythm raw-information of the sub-band with weighted or unweighted rhythm raw-information of the other sub-band to obtain the rhythm information of the audio signal.
8. Apparatus according to claim 7, wherein the means (114a) for deriving a weighting factor is disposed to derive a relative weighting factor for every sub-band signal, wherein a sum of the weighting factors for all sub-band signals equals 1.
9. Apparatus according to claim 8, wherein the means (114a) for deriving a weighting factor is disposed to derive a weighting factor as ratio of the significance measure of a sub-band signal to the sum of the significance measure of all sub-band signals.
10. Apparatus according to claim 9, wherein the means (106a, 106b) for examining a sub-band signal is disposed to examine a sub-band signal whose length is higher than 10 seconds.
11. Method for analyzing an audio signal with regard to rhythm information of the audio signal, comprising:

    dividing the audio signal into at least two sub-band signals (104a, 104b);

    examining (106a, 106b) a sub-band signal with regard to a periodicity in the sub-band signal to obtain rhythm raw-information (108a, 108b) for the sub-band signal;

    evaluating (110a, 110b) a quality of the periodicity of the rhythm raw-information (108a) of the sub-band signal (104a) to obtain a significance measure (112a) for the sub-band signal; and

    establishing the rhythm information of the audio signal under consideration of the significance measure (112a) of the sub-band signal and the rhythm raw-information (108a, 108b) of at least one sub-band signal.

Images