WO2010000807A2

WO2010000807A2 - A system and a method for configuring af loudspeaker system

Info

Publication number: WO2010000807A2
Application number: PCT/EP2009/058337
Authority: WO
Inventors: Morten Lydolf; Søren Borup JENSEN
Original assignee: Bang & Olufsen A/S
Priority date: 2008-07-03
Filing date: 2009-07-02
Publication date: 2010-01-07
Also published as: EP2308245A2; WO2010000807A3

Abstract

A system and a method for determining parameters for adapting a signal to be fed to each of a plurality of speakers where relations between a parameter of the speakers are determined and the adaptations determined on the basis of the characteristics of the individual speakers as well as the relations. The parameter may be an adaptation of the gain, phase or relative/absolute frequency responses of the speakers so as to achieve a high quality sound reproduction system with adapted speakers.

Description

A SYSTEM AND A METHOD FOR CONFIGURING A LOUDSPEAKER SYSTEM

The present invention relates to a system and a method for providing adaptation parameters for configuring a loudspeaker system and in particular to an improved manner of ensuring that multiple speakers for providing sound to a single room, space and/or compartment are adapted to each other.

When manufacturing speakers for buildings, cars or the like, such speakers will differ in characteristics due to imperfections or variations during manufacture. Most speakers operate by having a coil fed with a current move within a magnetic field (or vice versa), and the relative dimensions of these elements are important for the efficiency of the movement and thus of the response of the speaker to the signal fed. Also, other components of the speaker may vary and thus cause variations in the response. If the speaker is a dynamic speaker, the attached amplifier may also have varying characteristics due to e.g. manufacturing imperfections or variations.

US6674864 describes a system in which the signal provided to each of a multiple of speakers is adapted to that particular speaker using a FIR filter performing a compensation corresponding to the inverse of the speaker's response curve. Thus, the individual speakers have their sounds compensated to a predetermined sound. The speakers, thus, are individually compensated with no reference to the output of the others.

JP2003-274496 relates to a system of multiple speakers adapted to a space, such as the interior of a vehicle, where data describing adaptation of the sound of the individual speakers in relation to the vehicle interior is used for adapting the sound of the speakers. This adaptation is not related to the characteristics of the speakers themselves.

The present invention relates to improving high quality audio in e.g. vehicles by adapting the output of the individual speakers on the basis of both the characteristics of the individual speaker and/or amplifier as well as those of other speakers.

In a first aspect, the invention relates to a method of determining adaptation parameters for adapting a signal for each of a plurality of speakers, the method comprising:

1. accessing, for each of the plurality of speakers, one or more parameters relating to characteristics of the pertaining speaker,

2. for at least one first parameter of the one or more parameters, determining a relation between the first parameter(s) of the speakers, 3. determining the adaptation parameters for each speaker on the basis of the one or more parameters of the speaker and the relation.

In this respect, a speaker is an element adapted to output sound when fed a signal, such as an electrical signal. The speaker may comprise one or more sound generating units, such as units each being adapted to output sound within different frequency intervals (typically treble, bass and mid range). Also, the speaker may be active, i.e. comprise an amplifier adapted to amplify a received signal before feeding it to the actual sound generator, or not.

Parameters may be accessed by reading from a storage, internal or external or even remote, such as available over a LAN, a WAN, the WWW or a combination thereof. The parameters may be attached to or provided with the individual speakers or may be available from other sources, such as remote databases holding the parameters of vast numbers of speakers and which may be queried by the present method.

Multiple types of parameters characterize a speaker. Typically, these parameters characterize sound output by the speaker when fed with a known or standardized signal. Such parameters may be linear (such as gain and phase response) or non-linear (such as harmonic distortion and intermodulation distortion). These parameters may relate to a frequency response of the speaker, a phase thereof and/or a gain thereof. The frequency response relates to an amplitude of sound output at individual frequencies, normally when fed with a standard signal. The frequency response may be relative, which means that it describes how much stronger the output is at e.g. one frequency compared at another frequency, so that the frequency response of the individual speaker may be evaluated. Alternatively, the frequency response may be in absolute values which means that the frequency response is determined for a standard input, so that speakers may be compared to each other at different frequencies.

A difference in frequency response may be determined by one speaker providing certain frequencies or frequency intervals louder than other loudspeakers. Thus, when listening to a stereo with non-matched speakers, the voice of a vocal may, due to this imperfection, suddenly shift to one side in the stereo perspective when it changes frequency contents.

The gain may be determined in many manners, such as the absolute output of the speaker at one frequency when fed with a standardized signal. Alternatively, the output at a number of frequencies or within a frequency interval may be determined. The higher the overall output, the higher the gain. The gain may be more or less independent of the frequency response of the speaker. A difference in gain will be determinable by one speaker providing a louder sound (at some frequencies or in e.g. a frequency interval - or in general) than other speakers.

The phase describes the timing in the output of frequencies. A difference in the outputting time of a signal may be distracting to a listener, so it may be desired to also correct the phase of the speakers.

A difference in phase of two speakers may result in the correct stereo perspective being distorted, which means that the correct positioning of e.g. an instrument in the stereo perspective will be altered or blurred.

The parameters may relate to a speaker having a single sound provider or multiple sound providers and even comprising an amplifier. The parameters thus characterize the complete assembly.

Thus, the first parameter may be any parameter of the speakers. If the frequency response is chosen, the frequency response of the speakers may be adapted to each other so that the above disadvantages may be removed.

The relation selected may be a simple quantification, such as a difference in gain level and/or phase between two or more speakers. More complex situations may exist when comparing frequency response or phase differences at multiple frequencies, whereby the relation may require multiple values per pair of speakers. The relation normally is between the first parameter(s) of pairs of speakers but may also be between that of larger groups of speakers or between the individual speaker and a value representing the first parameter(s) of other speakers (such as the difference between the parameter of the speaker relative to a mean of the same parameter of all of or a group of speakers).

Having determined the relation, the parameters of the individual speaker may be adapted or added to using the relation. For example, if the first parameter is the gain, the gain parameter of the speaker may be adapted accordingly, so that the speaker has its parameter altered in accordance with the adaptation.

Alternatively, the relation may be added to the parameters of the speaker so that the parameters now characterizing the speaker in the present set-up is larger in number.

As will be described further below, these adaptation parameters may be used for adapting a signal output to the speaker in order for the sound output by the speaker to be adapted to the output (possibly adapted) of the other speakers. Preferably, step 2 comprises determining the relation on the basis of a first parameter representing a gain of the speakers, as the gain is considered to be the parameter experiencing the largest variation in speaker manufacture.

As mentioned above, the relation may relate to a difference of a parameter between the parameter of the individual speaker and a value derived from all speakers or the other speakers. Alternatively, step 2 comprises selecting a predetermined speaker and determining the relation as a relation between the first parameter(s) of the predetermined speaker and the first parameter(s) of the remaining speakers. Thus, the speaker with the largest or lowest value of the parameter may be selected and the others sought adapted to this extreme value. Alternatively, the speaker having a median/average value of the parameter may be selected to vary the parameter of all speakers as little as possible.

In one embodiment, step 2 is performed for each of two or more groups of speakers, the method further comprising the step of:

a) for at least one first parameter of the one or more parameters, determining a second relation between the first parameter(s) of the two or more groups of speakers,

wherein step 3 comprises determining the adaptation parameters for each speaker on the basis of the one or more parameters of the speaker, the relation of the pertaining group, and the second relation.

Thus, adaptation may be performed firstly within the individual groups and subsequently between groups. Naturally, adaptation within groups may be obtained using different parameters within the individual group. However, it is preferred that the same parameter is used both within groups and between groups.

A second aspect of the invention relates to a method of providing sound via a plurality of speakers, the method comprising :

- determining the adaptation parameters according to the first aspect of the invention,

receiving a signal or file representing the sound to be provided, generating a signal for each of the speakers by, for each speaker, generating a speaker signal from the received signal and the adaptation parameters for the pertaining speaker,

having the speakers generate the sound on the basis of the individual speaker signals.

As mentioned above, the adaptation parameters for each speaker relate to both the parameters of the speaker itself as well as the relation to other speakers.

The signal or file representing the sound may be a streamed signal received from the WWW or from a wireless connection (network, cellphone, radio waves or the like) or may be a file downloaded, accessed, stored, generated, purchased, leased or the like, from a storage medium, a network (LAN, WAN, WWW or the like) and which may be read and used for generating the signal for the speakers.

Usually, this signal may comprise individual signals for individual speakers or groups of speakers (right and left side of a stereo signal and the 5, 6, 7 or more channels of a surround sound system). Alternatively, or in addition, the signal may be divided up into different signals for providing to individual speakers (treble, bass, midrange).

These signals are then adapted on the basis of the adaptation parameters for the pertaining speakers in order to obtain the advantage of the invention.

Finally, the individual signals are fed to the speakers to which they are adapted. This feeding may be via cables, wireless or the like. Alternatively, or in addition, the signals may be stored for future use with the same speakers.

Also, the adaptation may be performed in real time, such as at the speaker, if it e.g. is an active speaker with its own amplifier. In that situation, the parameters may be fed to the amplifier, which then receives a standard signal (e.g. the received signal or part thereof not adapted to the individual speaker) and performs the adaptation before, after or while amplifying the signal and before feeding the, now adapted, signal to the speaker.

A third aspect of the invention relates to a system for determining adaptation parameters for adapting a signal for each of a plurality of speakers, the system comprising : means for accessing, for each of the plurality of speakers, one or more parameters relating to characteristics of the pertaining speaker,

first means for, for at least one first parameter of the one or more parameters, determining a relation between the first parameter(s) of the speakers, and

- second means for determining the adaptation parameters for each speaker on the basis of the one or more parameters of the speaker and the relation.

As mentioned above, a speaker may have one or more sound generating elements and may be active or not.

The accessing means may be adapted to read the parameters from one or more storage elements, such as portable storage elements, local or remote. Remote storage elements may be available over a network, such as a LAN, a WAN, the WWW, a cellphone network or the like.

The parameters may be any of the above parameters and may be represented in any desirable manner.

The first and second determining means normally will be one or more processors, hardwired or software controlled or a combination thereof, which may be positioned or available at or in the vicinity of the speakers or remotely therefrom. It is possible to feed the parameters to a remote processor performing the determinations and then simply receive the adaptation parameters for use locally with the selected speakers.

The configuring or adaptation process may be provided as a table with relations provided to be deduced by a solver to provide each of the values of the adjustment variables.

Naturally, the accessed parameters and/or the saved adaptation parameters or attributes as well as any key parameters may be loaded into the determining or configuring means using electronic means, such as means connected wirelessly or connected wired to e.g. an audio reproduction system. In one situation, this is facilitated by scanning of a bar code label or by manual means.

In one situation, the first determining means are adapted to determine the relation on the basis of a first parameter representing a gain of the speakers. In one embodiment, the first determining means are adapted to select a predetermined speaker and determine the relation as a relation between the first parameter(s) of the predetermined speaker relative to the first parameter(s) of the remaining speakers. This may be the maximum, minimum, median or mean parameter value as described above. Alternatively, the adaptation parameter may be determined in relation to the parameter of the pertaining speaker and a value determined from all other speakers or all speakers.

In one embodiment, the first determining means is adapted to determine the relation for speakers in each of two or more groups of speakers, the system further comprising means for, for at least one first parameter of the one or more parameters, determining a second relation between the first parameter(s) of the group(s) of speakers, wherein the second determining means is adapted to determine the adaptation parameters for each of the speakers on the basis of the one or more parameters of the speaker, the relation of the pertaining group and the second relation between the groups.

As mentioned above, even though it is preferred that the adaptation within the groups and between the groups is performed on the basis of the same parameter, this is not a requirement.

In a fourth aspect, the invention relates to a system for providing sound from a plurality of speakers, the system comprising:

the system according to the third aspect of the invention,

- means for receiving a signal or file representing the sound to be provided,

means for generating a signal for each of the speakers by, for each speaker, generating a speaker signal from the received signal and the adaptation parameters for the pertaining speaker.

As mentioned above, the signal or file may be received from a wireless connection, WiFi, radio waves, cellphone or the like, a local or remote storage medium, via a network or the like and as a stream or a complete file.

The signal generating means normally will be able to, from the signal or file, generate a plurality of signals each for a speaker or a group of speakers, as is usual in e.g. stereo amplifiers. The signals may be split up further to individual speakers as is seen in speaker systems with multiple sound generators (treble, bass, mid range). Thus, this may be a multiple step process, and the generating means may be a single means or means distributed within the system.

The adaptation parameters may be used centrally, or standard signals (such as the two channels of a stereo signal and the multiple channels of a surround sound system) may be forwarded to distributed means for later adaptation.

Actually, the speakers (or some of them) may be active speakers comprising an amplifier, and the adaptation parameters may be fed to the such amplifiers for adaptation of the signal received by the amplifier before feeding the amplified and adapted signal the pertaining speaker. In this manner, a standard element may be used for receiving the file/signal and dividing it into a number of signals to be received by e.g. active speakers. Such active speakers are then provided with adaptation parameters and adapted to perform this adaptation before feeding the signal to the speaker.

Thus, the determining/configuring means may be embedded into an audio reproduction system such as an audio reproduction system comprising two or more active sound transducers, the system to constitute an audio reproduction equipment in a room, in a car, in a boat and alike.

In one embodiment, the system further comprises a plurality of speakers each being connected to the generating means for receiving the signals and outputting the sound. Naturally, this connection may be wireless or via cables. Alternatively, the adapted signals may be stored for future use on a standard sound producing system based on the speakers.

In the following, preferred embodiments will be described with reference to the drawing, wherein:

Figure 1 illustrates components that constitute examples of loudspeaker module configurations;

- Figure 2 illustrates an example of an embodiment of the invention;

Figure 3 illustrates a system block diagram of one embodiment of the configuring means.

Figure 1 illustrates components that constitute examples of loudspeaker module configurations. A loudspeaker transducer unit 1 is the fundamental means that transforms the electrical signal to the sound waves produced by movements of the membrane of the transducer unit

Any standard transducer unit may be applied in the present embodiments.

The characteristics of each loudspeaker transducer unit may be determined by measurement and/or from specifications delivered by the supplier of the unit.

In a preferred embodiment the characteristics to be applied in the configuring process are:

Gain;

Frequency response (absolute or relative)

Phase response

- Harmonic distortion

Intermodular distortion.

In an audio reproduction system comprising active sound transducers an amplifier 2 is provided for each transducer unit 1. This type of amplifier is e.g. the technology ICEpower from Bang & Olufsen DK.

The transducer unit 1 and the amplifier 2 may constitute a loudspeaker module, which may be characterized as a module having attributes for gain and frequency response. Thus the configuring procedure may operate on attributes associated with the transducer unit 1 itself, or it may operate on attributes associated with the loudspeaker module comprising both the transducer unit 1 and the amplifier 2.

In a high quality audio reproduction system, a dedicated filter means 3 is provided per amplifier 2. The means (3) provides a frequency dependent amplification to control the overall gain, which may be regulated up or down as required. Means for down regulation may be as simple as adjustment of a resistive means serial connected to the loudspeaker module.

Figure 2 illustrates an example of an audio reproduction system comprising loudspeaker modules configured in a stereophonic concept with two LEFT channel loudspeaker modules 22 and with two RIGHT channel loudspeaker modules 23. The audio reproduction system may be an audio amplifier 20 controlling and driving the loudspeaker modules 22,23, on the basis of a received signal 21.

In a preferred embodiment, the measured electro acoustical attributes per loudspeaker transducer unit is a subset, e.g. but not limited to, frequency response and gain. The attributes may be represented as a data set like a table.

The relations that define an actual configuration contain relations among loudspeaker system topology and speaker module attributes.

In the following, examples of the present invention are given using the gain, even though these could have been made using an adaptation of the relative frequency response and/or phase.

A speaker system topology may define the configuring strategies like:

apply the loudspeaker having the maximum gain as a reference unit;

apply the loudspeaker having the minimum gain as a reference unit;

apply the loudspeaker having an average gain as a reference unit ;

apply an all-to-all speaker validation adjustment;

apply a group wise speaker validation adjustment, i.e. validate speakers that is members of a defined group of speakers;

apply a group-to-group validation adjustment; apply an adjustment within one or more groups.

A constraint table is used for defining the speaker system concept, the configuring strategy, and the operators to apply accordingly. It is a great advantage to have a table to define and control the configuring procedure:

- the table is a data set representing one or more specific loudspeaker modules constituting the end user system;

the table may follow the physical package of speaker modules to be configured;

the loudspeaker configuring means may be generic, as it only becomes specific when a specific table is loaded by the controller of the audio sound system so as to finally control the two or more loudspeakers.

An example of the operators offered in the table in a preferred embodiment is:

Strategy: maximum, minimum, average, predefined;

Concept: speaker ID#, type (left, right, centre,), concept (group, surround, other);

Relation: greater than, less that, equal, add, subtract (>, <, =, +, -).

In the following, the notation applied is: Left-x is a left channel speaker module; Left-y is another left channel speaker module; Right-x is a right channel speaker module, Right-y is another right channel speaker module;

Thus, an all-to-all adjustment using the maximum gain of a system having four speakers, two left and two right, could be:

- identify the speaker with the maximum gain(max of: Left-x, Left-y, Right-x, Right-y);

apply the identified maximum gain value to the other speakers (identify a difference and use this) and adjust the other speakers accordingly: Left-x, Left-y, Right-x, Right-y = maximum gain.

This could be represented in a table format providing the desired information in a compact manner:

In this table, the right and left speakers are defined as groups, but all four speakers are adjusted on an all-to-all basis.

In another example, the four speakers are divided into two groups, two in one "left" group and two in a "right" group. Again, all speakers may be adjusted on an all-to-all basis, and subsequently, it is desired that the left speakers are further adjusted with a delta gain (so as to have the left side play louder than the right side). In this example, we choose to apply the minimum gain as reference:

identify the minimum gain(min of: Left-x, Left-y, Right-x, Right-y);

apply the identified minimum gain value and adjust: Left-x, Right x = minimum gain; Left-y, Right-y = minimum gain + delta gain;

or on the table notation:

A next example relates to a situation again having four speakers, two left and two right, which are all to have the same, predefined gain:

apply the predefined gain value and adjust accordingly to all speakers: Left-x, Left-y, Right-x, Right-y = predefined gain.

This method may be performed independently of the future use of the loudspeakers or may form part of the actual audio reproduction system driving the loudspeakers when in use. Thus the system then becomes the controller of the configuring process that provides a configuration of the system itself.

The saved attributes and key parameters are loaded into the configuring means supported by electronic means, means that is connected wirelessly or connected via wires to the audio reproduction system. The saved attributes may be saved on any type of digital media e.g. a server, SD card, PROM, EPROM, SRAM and alike, and/or as a bar code label to be scanned. Alternatively to an electronically controlled upload of the required dataset or algorithms the saved attributes and key parameters may be loaded into the configuring means supported by manually means by entering data.

As mentioned, the loudspeakers may be divided into groups and adapted within the groups and between groups, such as if a left to right calibration is performed :

calibrate a LEFT channel in relation to a RIGHT channel so the gain difference between the two channels is less than a given maximum, i.e. ABS (gain left (x) - gain right (x)) <= xa dB; calibrate the two LEFT channels in relation to each other so the gain difference between the two channels is less than a given maximum, i.e. ABS (gain left (x) - gain left (y)) < = xb dB;

calibrate the two RIGHT channels in relation to each other so the gain difference between the two channels is less than a given maximum, i.e. ABS (gain right (x) - gain right (y)) <= xb dB;

Any number of groups may be used, such as 5, 6, 7 or more in surround sound systems, 2 in stereo systems or the like.

Figure 3 illustrates a preferred embodiment of an automatic configuring means 30 for performing the determinations. This means 30 may form part of a stereo of e.g. a vehicle or may simply be used for providing parameters for use with the loudspeakers in another setup.

The means 30 comprises a storage 37 storing attributes from the one or more loudspeaker transducer units, which alternatively may be accessible from a remote link, and which are available over an I/O port 35.

A storage 36 holds configuration relations defining what's valid on a specific set of speaker attribute values. These relations constitute the solution space with legal combinations of the loudspeaker adjustment parameters.

Configure means 31 loads the Configuration relations 36 and with the Attribute values 35 as input, it deduces the legal solution that's applicable for the actual defined loudspeaker speaker concept. The result of the deduction is an updated list of Adjustment parameter values 34 to be applied in the adjustment of the loudspeaker modules 38.

The configure means is imbedded into the controller of the audio sound system, which is the master having the Adjust means 32 that initiates controls and applies the configuring and calibration process. This Adjust means 32 may be used by a user or listener for altering characteristics of the sound (more treble, bass, louder right or left side or the like).

The invention includes transducer configurations to be applied in surround sound systems and traditional stereophonic system, and to be applied in any domain like a home, in a car, in a boat, in an office or in any other public domain.

Claims

1. A method of determining adaptation parameters for adapting a signal for each of a plurality of speakers, the method comprising:

2. for at least one first parameter of the one or more parameters, determining a relation between the first parameter(s) of the speakers,

3. determining the adaptation parameters for each speaker on the basis of the one or more parameters of the speaker and the relation.

2. A method according to claim 1, wherein step 2 comprises determining the relation on the basis of a first parameter representing a gain of the speakers.

3. A method according to claim 1 or 2, wherein step 2 comprises selecting a predetermined speaker and determining the relation as a relation between the first parameter(s) of the predetermined speaker and the first parameter(s) of the remaining speakers.

4. A method according to any of claims 1-3, wherein step 2 is performed for each of two or more groups of speakers, the method further comprising the step of:

5. A method of providing sound via a plurality of speakers, the method comprising:

determining the adaptation parameters according to any of claims 1-4,

6. A system for determining adaptation parameters for adapting a signal for each of a plurality of speakers, the system comprising :

means for accessing, for each of the plurality of speakers, one or more parameters relating to characteristics of the pertaining speaker,

- first means for, for at least one first parameter of the one or more parameters, determining a relation between the first parameter(s) of the speakers, and

second means for determining the adaptation parameters for each speaker on the basis of the one or more parameters of the speaker and the relation.

7. A system according to claim 6, wherein the first determining means are adapted to determine the relation on the basis of a first parameter representing a gain of the speakers.

8. A system according to claim 6 or 7, wherein the first determining means are adapted to select a predetermined speaker and determine the relation as a relation between the first parameter(s) of the predetermined speaker relative to the first parameter(s) of the remaining speaker(s).

9. A system according to any of claims 6-8, wherein the first determining means is adapted to determine the relation for speakers in each of two or more groups of speakers, the system further comprising means for, for at least one first parameter of the one or more parameters, determining a second relation between the first parameter(s) of the groups of speakers,

wherein the second determining means is adapted to determine the adaptation parameters for each of the speakers on the basis of the one or more parameters of the speaker, the relation of the pertaining group and the second relation between the groups.

10. A system for providing sound from a plurality of speakers, the system comprising : the system according to any of claims 6-9,

means for receiving a signal or file representing the sound to be provided,

11. A system according to claim 10, the system further comprising a plurality of speakers each being connected to the generating means for receiving the signals and outputting the sound.

12. A system according to claim 10 or 11, wherein the receiving means comprise means for reading a storage medium comprising the file.