EP1883064B1 - Musical instrument with sound transducer - Google Patents

Description

The invention relates to a sound transducer, which transforms an excitation signal generated by at least one resonator into a sound signal, and in which the sound transducer is provided with an adjustable oscillation profile, wherein at least one profile parameter is defined by a reference profile of a reference instrument, and a corresponding method for adaptation a sound transducer.

The sound production of a musical instrument is usually done by the interaction of three individual components. Initially, an excitation takes place, with which an excitation energy is entered into the instrument. This can be for example the bow of a violin, the hammer of a piano or the mouthpiece and the leaflet of a saxophone. Another component concerns the presence of one or more resonators which determine the fundamental frequency of the individual tones. The resonators can have constant or variable properties. Resonators are, for example, the strings of a violin or a guitar or the variable air column of a saxophone. The excitation energy introduced into the resonator, together with the exciting element and the resonator, leads to the generation of sound.

As a third component acts on the sound generator with a sound transducer or sound box, which transforms the vibration energy provided by the instrument in sound level in the surrounding air and the vibration energy transported thereby the highest possible efficiency of the instrument in the air. Such resonant bodies are for example the body of a violin, the funnel of a saxophone, the pickup of an electric guitar or the loudspeaker of a guitar amplifier.

The sounder or sounder typically represents an acoustic or electroacoustic impedance transducer and is typically designed to achieve a high perceived loudness of the instrument. In addition, the sound transducer or resonator fulfills the task of shaping the frequency spectrum of the instrument, thereby producing a pleasant and characterful sound of the instrument.

The tonal character influenced by the resonant body essentially determines the perceived quality of a musical instrument and thus also the quality of a musical lecture. The acoustic properties of the transducer are determined both by its geometry and by the materials used.

In the acoustic sense, the individual sound character of an instrument is determined by the so-called formants. These formants are narrowband increases in the frequency spectrum and usually independent of the pitch played. The human ear responds very sensitively based on a pattern recognition on these formants, so that even musically untrained people, for example, can easily distinguish the sound of a violin from the sound of a viola, although the instruments differ essentially only by the size of your body.

Due to a variety of physical parameters that determine the sound of a particular transducer or the sound box, it usually proves to be extremely expensive to provide a sound box with a precisely predetermined sound profile. Another major problem is when a larger number of resonant bodies are to be produced with substantially the same predetermined sound images.

From the DE 103 92 940 T5 It is known to modify a generated by a transducer of a musical instrument sound signal. The transducer is provided with an adjustable vibration profile. Modification of the vibration signal is done using a signal processor.

The US 2005/0045027 A1 describes a variable memory for frequency responses to be generated to match the sound of a musical instrument to other given instruments. Corresponding stored frequency responses are retrieved and processed as part of a control.

In the US 6,740,805 B2 A method for broadcasting a previously acquired and stored spatial sound event is described. In a first step, the sound event is recorded via several microphones distributed in the room and Subsequently, the recording is reproduced without the inclusion of another musical instrument.

From the DE 20 2004 008 347 U1 is a method for the algorithmic generation of melodies taking into account external influencing factors known.

In the US 5,578,548 A A method of processing is described to summarize the frequency response of a musical instrument's sounding body and the frequency response of an excitation of the resonator. The summary can reduce the required processing power and storage capacity.

From the US 6,392,135 B1 For example, a virtual musical instrument is known which can reproduce stored sound material in an adaptable manner.

In the US 2004/187673 A1 An apparatus for transforming fundamental vibrations of the strings of an untuned instrument into fundamental vibrations of the strings of a tuned instrument is described. For this purpose, the actual fundamental oscillations are detected by measurement technology and shifted to suitable values in the context of digital data processing.

In the US B1-6,448,488 An electromechanical pick-up system of a stringed instrument is described.

The US B1-6,504,935 illustrates a method for analyzing and replicating nonlinearity according to a given nonlinearity.

From the US-A 5,133,014 is a semiconductor-specific simulation of the transmission characteristics of a tube amplifier, which is intended for use in connection with electric guitars.

The object of the present invention is to design a sound transducer such that the generation of a predetermined sound image is supported, and to provide a method for adapting a sound transducer to a given sound image.

This object is achieved by the method according to claim 1 and the sound transducer according to claim 2. Advantageous embodiments of the sound transducer are specified in the dependent on the patent claim 2 claims 3 to 5.

By designing the sound transducer with an adjustable vibration profile and by the parameterization of this vibration profile as a function of the reference profile of a reference instrument, it is possible to model the reference profile of any reference instrument to be selected. It is thus possible to approximate the sound image of the sound transducer largely to the sound image of the reference instrument. In particular, it is also possible to produce a large number of sound transducers with the same sound patterns determined by the reference profile of the reference instrument. In particular, it is thought that the sound transducer generates a sound signal as a sound transducer.

A realization of the sound transducer is typically such that the sound transducer is designed as a sound box.

One embodiment is that the sound transducer is designed as an acoustic sound box.

In addition, it is also thought that the sound transducer is formed as an electroacoustic sounding body.

In order to provide a characteristic reference profile, it is proposed that the reference profile is determined by mean value formation taking place during a predefinable time interval.

A meaningful storage of a sound image can take place in that the reference profile is defined by the frequency response of the reference instrument.

In particular, a precise definition of the sound image is supported by the fact that the reference profile defines a statistical distribution of pitches and volumes.

Immediate detection of a reference profile can take place in that the reference profile is defined by evaluation of music played at the time of evaluation.

An objective presence of the reference instrument can be dispensed with by defining the reference profile by evaluating recorded music.

An adaptive frequency response adaptation of the sound transducer is supported by the fact that a measuring device for measuring a frequency response of the sound transducer is coupled to the sound transducer.

A concrete realization of the parameterization takes place in that the sound transducer has at least one adaptive element for frequency response adaptation.

An exact adaptation of the intrinsic profile to the reference profile can take place in that the sound transducer is designed as part of a control loop for evaluating a difference between the reference profile and an intrinsic profile of the sound transducer.

An individual influenceability is achieved in that a parameterization of the sound transducer can be influenced manually.

Fig. 1

eine schematische Darstellung zur Erzeugung, Erfassung und Verwendung eines Referenzprofils zur Parametrisierung eines Klangwandlers,

Fig. 2:

eine schematische Darstellung zur Veranschaulichung einer Veränderung eines Eigenprofils eines Klangwandlers unter Berücksichtigung eines Referenzprofiles und

Fig. 3:

eine schematische Darstellung zur Veranschaulichung einer Ausführungsvariante, bei der die Übertragungscharakteristik eines Klangwandlers an die Übertragungscharakteristik eines Referenzklangwandlers angepasst wird.

In the drawings show:

Fig. 1

a schematic representation of the generation, detection and use of a reference profile for the parameterization of a sound transducer,

Fig. 2:

a schematic representation for illustrating a change of an intrinsic profile of a sound transducer taking into account a reference profile and

3:

a schematic representation for illustrating an embodiment in which the transmission characteristic of a sound transducer is adapted to the transmission characteristic of a reference sound transducer.

Fig. 1 shows in a schematic block diagram representation the generation, storage and use of a reference profile (1). From a reference instrument (2), which is formed from a sound generator (3) and a reference sound transducer (4), a reference sound is generated directly or by using a loudspeaker (5), which is detected by a microphone (6). The microphone (6) is connected to a reference memory (7), which allows storage of the reference profile. The reference memory (7) is coupled to a signal processor (8) which in particular supports a statistical evaluation of the sound image captured by the microphone (6).

A detection of the reference profile (1) can for example be such that a sufficiently long musical performance on a specific reference instrument (2) detected and using the signal processing (8) statistically with respect to the characteristic frequency response of the reference instrument (2) and its reference sound transducer (4 ) is evaluated. The signal processing (8) may include averaging.

By the signal processing (8) not only a frequency response of the reference sound transducer (4) is determined, but it is also the frequency response of Sound production analyzed and recorded. The result of the statistical evaluation is therefore also dependent on the type of musical lecture and in particular on the statistical distribution of the pitches played and their volume. A typical reference profile thus contains for individual frequency components the respective amplitude values or the relative amplitude components over the entire signal amplitude. A quantization of the frequency response is carried out with a sufficiently fine subdivision.

Due to the coupling of the microphone (6) to the reference memory (7) or the signal processing (8), it is not necessary for the reference instrument to be physically present during the execution of the signal processing. An audio recording of the sound image of the reference instrument (2) proves to be sufficient. According to one embodiment of the invention, it is also particularly conceivable to deposit a plurality of different reference profiles (1) in the region of the reference memory (7). A user can thus select between several reference profiles (1).

According to the embodiment in FIG Fig. 1 the musical instrument (9) a sound generator (10) which is coupled to a sound transducer (11). The sound transducer (11) generates an acoustic signal which is supplied to an environment directly or by using a loudspeaker (12). A current intrinsic profile (13) of the sound transducer (11) is fed to a difference formation (14), which evaluates the reference profile (1) as the second input variable. The output signal provided by the difference formation (14) is supplied to the sound transducer (11), taking into account a gain (15), and here parameterizes its concrete present sound pattern.

In the event that the difference formation (14) delivers the value zero as the output signal, the sound transducer (11) has its sound image varied in such a way that the current intrinsic profile is equal to the predetermined reference profile (1). The sound transducer (11) is typically designed such that it has a variable and parameterizable sound image and continuously measures the frequency response of its output signal during the musical performance. The sound transducer (11) thus automatically determines its own profile (13) at the same time to generate the musical lecture.

By a permanent or cyclic comparison of the intrinsic profile (13) with the reference profile (1), a change of the variable sound transducer (11) takes place in such a way that the differences between the intrinsic profile (13) and the reference profile (1) are minimized. The musical instrument (9) thus adapts the sound image of the reference instrument (2) adaptively by the parameterization of its sound transducer (11).

The frequency response adjustment of the musical instrument (9) can be done automatically or interactively with a user. In particular, it is possible to carry out a manual influencing of the adaptation process such that the user can interactively control the frequency response adaptation by the nature of his musical presentation. In particular, the musician can control the approach to the reference profile (1) by the statistical choice of pitches and volumes.

In addition, it is possible that the artist in artistic intention by a conscious different kind of game of the musical instrument to be parametrized (9) the behavior of the sound transducer (11) away from the reference profile (1), thereby generating an individual sound.

The above explanations on the construction of the sound transducer (11) and the associated functional components in combination with the musical instrument apply in the same way even in a device implementation without assignment to a musical instrument and without simultaneous generation of an acoustic sound signal. The loop provided by the feedback comprises according to the embodiment in FIG Fig. 1 the airway between the speaker (12) and the microphone (6).

Fig. 2 Fig. 12 illustrates a process in which the reference profile (1) is generated using an audio recording stored on a cassette (16), for example. The procedure corresponds essentially to the representation in FIG Fig. 1 , In addition, an ear (17) of a user or a listener is shown. It is also possible to use reference profiles (1) stored in a different manner.

As an alternative to a direct generation of an acoustic sound signal by the sound transducer (11), it is also possible that the sound transducer (11) is an electrical or other output signal generated, which is converted in a further processing step at the same time or offset in time into an acoustic sound signal. For example, the immediate output signal of the sound transducer (11), which is generally referred to as a sound transducer in the following text, is first recorded and transformed into audible sound after a corresponding storage at a later time.

The sound transducer (11) can also be realized as a digital or analog circuit whose output signal is fed to an amplifier or directly to a loudspeaker or to a different type of sound generator. In a digital realization of the sound transducer (11), in particular, it is also intended to carry out the signal processing using a Fourier transformation.

In another embodiment, the sound transducer (11) may be implemented as an adaptive filter. According to the embodiment in FIG Fig. 2 it is not mandatory that an acoustic airway is included as a transmission link in the provided control loop.

Fig. 3 shows an embodiment in which the sound generator (10) of the musical instrument (9) supplies its output signal to both the sound transducer (11) and the reference sound transducer (4). The output signal of the sound transducer (11) is in turn fed back via the intrinsic profile (13), the difference formation (14) and an adjustable gain (15). On the difference formation (14), the output signal of the reference sound transducer (4) with the interposition of the reference memory (7) and the signal processing (8) is also performed. The transmission profile of the sound transducer (11) can thereby be adapted to the transmission profile of the reference sound transducer (4). In particular, this can be achieved by adapting a sound converter (11), which has comparatively inexpensive design, with regard to its transmission behavior to the transmission behavior of a high-quality reference sound transducer (4).

The adaptation of the transmission behavior of the sound transducer (11) according to the embodiment in Fig. 3 can at a simultaneous supply of the output signal of the tone generator (10) both to the reference sound transducer (4) and to the sound transducer (11), but it is also possible, initially offset in time using the reference sound transducer (4) to store the reference profile (1) and to temporally later process steps adapt an arbitrary number of transducers (11) to the reference profile (1). The method is therefore also suitable for carrying out a series production. The series production can take place both by an individual adaptation made for each device and by using a once determined adaptation profile which is stored and used identically for each device to be adapted.

Previously, the sound transducer (11) used was preferably explained as a sound transducer. However, the actual generation of sound and / or the processing of an initially present as a sound signal input signal is not an indispensable part of the invention. Rather, the described transducer is only one embodiment of the sound transducer. The transducer can also be implemented as a speaker, linear or non-linear amplifier, guitar amplifier, processor or audio effect processor. The implementation can be done either analog, digital or partially analog and partially digital. Use as a sound transducer can also find signal processors.

The evaluated reference sound profile can be determined acoustically via the already explained reference sound transducer (4), but it is also a purely electronic processing conceivable. When evaluating a sound profile of an actual instrument, it is possible to evaluate the already mentioned musical lecture on this instrument, but it is also possible to deliberately subject the instrument mechanically or electrically to an excitation function and to analyze the corresponding output signal. It is not necessarily musical sounds generated in the true sense, but the sound production can be done in response to test signals or test suggestions.

The sound transducer and the reference sound transducer are not necessarily an inseparable part of a musical instrument and may be stimulated to perform the measurements both with a musical instrument and with a different analytical broadband signal.

Alternatively or additionally, it is also possible to use sound transducers or sound transducers which have a non-linear transfer function. The difference to linear sound converters is that the generated spectrum of the sound transducer is dependent on the amplitude of the input signal. In addition, a polyphonic musical instrument generates intermodulations or distortions and overtones in the nonlinear sound transducer.

These nonlinearities are often intentional and considered to be part of the sonic nature of the transducer. An example is a guitar amp or a speaker, or a combination of both. The amplifier is often operated in the non-linear range, in which the sound transducer (the loudspeaker) generates distortions in the amplifier output stage due to its high energy consumption. The loudspeaker itself also produces a high harmonic distortion factor because the damping diaphragm suspension gets out of its linear range with large signal deflections.

Also other z.T. Historical sound processors such as analog equalizers can be used here. They produce nonlinearities that have a positive effect on the sound in addition to the frequency response change.

Typical nonlinearities limit the signal amplitude up or down. This is done more or less gently depending on the characteristic. Small amplitudes remain nearly linear and uncompressed.

It turns out that a nonlinear sound transducer can be broken down into three components: the pure nonlinearity, and the frequency responses before and after this nonlinearity.

At high signal levels, the input frequency response primarily determines the character of the distortion and intermodulation. The output frequency response, however, generates the characteristic formants of the sound transducer. At low signal levels, nonlinearity has no significance and can be neglected. Here, both frequency responses are perceived as a single frequency response.

Based on the device of the adaptive sound converter will be explained below a device which detects both frequency responses of a nonlinear reference sound transducer and applied to an adaptive non-linear sound transducer.

In particular, the sound transducer has two separate vibration profiles with an intermediate nonlinearity.

• Ein Referenzprofil A bei geringem Eingangspegel. Hier spielt die Nichtlinearität der Referenz für den Frequenzgang keine Rolle. Dieses erste Profil repräsentiert die Multiplikation beider Frequenzgänge. Der Gesamtpegel wird durch die Verstärkung der Nichtlinearität um ihren Nullpunkt bestimmt.
• Ein zweites Referenzprofil B bei hohem Eingangspegel. Hier grenzt die Nichtlinearität beide Frequenzgänge voneinander ab. Die nun entstehenden Intermodulationen und Obertöne werden ausschließlich von dem vorgelagerten Frequenzgang bestimmt. Das aus der Nichtlinearität resultierende Frequenzspektrum wird vom nachgelagerten Frequenzgang geformt. Der Gesamtpegel wird durch die absolute Amplitudenbegrenzung der Nichtlinearität bestimmt.

The reference sound transducer now determines two reference profiles:

• A reference profile A at a low input level. Here the nonlinearity of the reference for the frequency response is irrelevant. This first profile represents the multiplication of both frequency responses. The overall level is determined by the gain of the non-linearity around its zero point.
• A second reference profile B at a high input level. Here the non-linearity delimits both frequency responses. The resulting intermodulation and overtones are determined exclusively by the upstream frequency response. The frequency spectrum resulting from the nonlinearity is shaped by the downstream frequency response. The overall level is determined by the absolute amplitude limitation of the nonlinearity.

• In der ersten Stufe wird der adaptive Klangwandler als linear angenommen mit einem Frequenzgang L. Bei geringem Eingangspegel wird dieser Frequenzgang L derart geregelt, dass sein Eigenprofil dem Referenzprofil A entspricht. Dieser Vorgang entspricht exakt dem zuvor beschriebenen Regelkreis. Der ermittelte Frequenzgang L ist jedoch lediglich ein Zwischenergebnis: Er entspricht der Multiplikation der beiden Frequenzgänge vor und nach der Nichtlinearität. Der individuelle Verlauf der Frequenzgänge ist jedoch noch nicht bekannt.

The adaptive nonlinear sound transducer is preferably adapted in two stages to the reference sound transducer:

• In the first stage, the adaptive sound transducer is assumed to be linear with a frequency response L. At a low input level, this frequency response L is controlled such that its intrinsic profile corresponds to the reference profile A. This process corresponds exactly to the previously described control loop. However, the determined frequency response L is only an intermediate result: it corresponds to the multiplication of the two frequency responses before and after the nonlinearity. The individual course of the frequency responses is not yet known.

In the second stage, the adaptive sound transducer is switched as the described combination of two frequency responses A and B with an intermediate nonlinearity. At high input level, the frequency response B is controlled such that its own profile corresponds to the reference profile B. This process corresponds exactly to the previously described control loop.

However, the intrinsic profile is additionally influenced by the frequency response A and the nonlinearity. Here, the control loop experiences its second feedback: while frequency response B is controlled, frequency response A is simultaneously modified in such a way that the multiplication of frequency response A and B corresponds to the previously determined frequency response L.

Therefore, frequency response A is controlled inversely: If a spectral component of frequency response B is increased in level, the corresponding spectral component of frequency response A is lowered in the same mass. Thus, the combined serial frequency response L is maintained.

Frequency response A also has an influence on the intrinsic profile of the sound transducer, and thus on the control, despite the downstream non-linearity. However, due to the compressive effect of nonlinearity, this influence is less than the influence of frequency response B. This guarantees that the control process is not unstable or indifferent at any point.

If the difference between reference profile B and the intrinsic profile was minimized in the second stage via the control process, then the frequency responses A and B are exactly approximated.

In addition to the frequency responses A and B, the character of the intermediate non-linearity has a decisive influence on the dynamic sound behavior of the sound transducer.

The present invention relies essentially on a trivial nonlinearity, as occurs throughout nature.

• Ein quasilineares Verhalten bei geringer Amplitude
• Eine absolute Amplitudenbegrenzung oben und unten.
• Ein monotoner Kennlinienverlauf
• Keine Hysterese
• Kein Gedächtnis: Die Nichtlinearität liefert für dieselbe Eingangsgröße immer dieselbe Ausgangsgröße, unabhängig vom bisherigen Signalverlauf.

The conditions of trivial nonlinearity are:

• Quasilinear behavior at low amplitude
• An absolute amplitude limit above and below.
• A monotone characteristic curve
• No hysteresis
• No memory: Nonlinearity always returns the same output for the same input, regardless of the signal history so far.

The trivial nonlinearity has two fundamental parameters: the quasi-linear gain and the absolute amplitude limit.

These two parameters can be chosen freely in the non-linear sound converter according to the invention. They are detected by the described two-stage determination of the intrinsic profile in the frequency responses L and A and B and compensated via the control process, since all frequency responses and profiles inherently include absolute level gains or attenuations.

The combined frequency response L thus corrects the gain in the quasi-linear range. The downstream frequency response B corrects the level of absolute amplitude limitation of the non-linearity.

Claims (5)

1. Method for adapting an acoustic transducer to a reference acoustic transducer,
   wherein the acoustic transducer comprises a first oscillation profile with a first frequency response, a second oscillation profile with a second frequency response and a trivial non-linearity inserted between the first and second oscillation profile,
   and wherein the acoustic transducer has its own characteristic profile, which corresponds to the frequency response, which results from the combination of the first oscillation profile, the trivial non-linearity and the second oscillation profile, comprising
   determining a first reference profile of the reference acoustic transducer at a low input level and a second reference profile of the reference acoustic transducer at a high input level;
   controlling the first and second frequency response of the acoustic transducer such that at a low input level, at which its own characteristic profile is not influenced by the trivial non-linearity, the characteristic profile corresponds to the product of the first and second frequency response; and
   controlling the first reference profile of the reference acoustic transducer such that at a high input level, at which its own characteristic profile is influenced by the trivial non-linearity, the difference between the characteristic profile and the second reference profile is minimized and the product of the first and second frequency response still corresponds to the first reference profile.
2. Acoustic transducer,
   wherein the acoustic transducer comprises a first oscillation profile with a first frequency response, a second oscillation profile with a second frequency response and a trivial nonlinearity inserted between the first and second oscillation profile,
   wherein the acoustic transducer has its own characteristic profile, which corresponds to the frequency response, which results from the combination of the first oscillation profile, the trivial non-linearity and the second oscillation profile, and
   wherein the acoustic transducer is adapted according to the method pursuant to claim 1.
3. Acoustic transducer according to claim 2 characterized in that the acoustic transducer is designed at least in part as a digital circuit.
4. Acoustic transducer according to any one of claims 2 or 3 characterized in that the acoustic transducer comprises a device for conducting a Fourier transformation.
5. Acoustic transducer according to any one of claims 2 to 4 characterized in that the acoustic transducer is designed as part of a digital guitar amplifier.

Images