EP0776144A1 - Signal modification circuit - Google Patents

Abstract

The circuit is used for a pair of signals (L,R) provided by a signal source (q), which are divided into signal components (s1-s6), and selectively combined to provide a pair of modified signals (L',R'). A first combining stage (K1) combines two signal components (s1,s2) obtained from the first signal (L), provided by the source, with a third signal (s3) obtained from the second signal, provided by the source. A second combining stage (K2) similarly combines two signal components (s4,s5), obtained from the second signal, with a signal (s6) obtained from the first signal.

Description

The invention relates to an electronic circuit for modifying a first and a second signal, which are either present individually or in connection with other signals. Such circuits are known to amplify or attenuate certain effects of the information contained in the signals. One application is, for example, a contour amplifier for signals which are linked to optical signals and which are formed by raster scanning or by a large number of sensors. There are similar applications for signals associated with sound waves, which range from very low frequencies to far into the ultrasound range. These signals can be used to detect seismic signals, but also high-frequency signals in the ultrasound range, as are used, for example, in material testing. Included is the normal audio area, which deals with audible signals. In the simplest case, the modification circuit relates to stereo signals according to one of the standardized stereo coding methods, which transmit a left and a right signal in coded form as a sum and difference signal. The so-called stereo basis can be changed electronically by means of the known modification circuits, as a result of which the two associated loudspeakers move apart as it were. The change in effects can also relate to more complex reproduction systems with more than two loudspeakers and / or more than two signals which convey a surround sound which can be changed by the modification circuit.

From US 5,136,650, for example, a complex circuit arrangement for sound signals is known in which six spatially distributed loudspeakers are individually controlled, originally only two signals being assumed. An effect control simulates a spatial sound that was not originally available.

From the magazine "Elrad", 1994, Issue 7, pages 76 to 81, under the title "Effekthascherei" it is described how electronic means broaden the stereo basis for conventional stereo signals and finally a spatial effect using a surround matrix can be achieved. Three basic circuits are presented, each of which is optimal for certain signal properties. Since the circuits cannot follow changed signal properties, their use is problematic, because in these cases it is possible that the effect control leads to an overall deteriorated hearing impression.

The object of the invention is to provide a circuit arrangement for modifying at least two signals, which can be easily adapted to the respective signal properties.

• Schaltung zur Modifikation eines ersten und zweiten Signals von einer, mindestens zwei Signale liefernden Signalquelle mit Einrichtungen zur Bildung von Signalkomponenten, die aus dem ersten und zweiten Signal gebildet und mittels einer ersten und zweiten Kombinationseinrichtung zu einem modifizierten ersten und modifizierten zweiten Signal zusammengefaßt sind, wobei
• in der Modifikationsschaltung eine mit dem ersten Signal verkoppelte erste und zweite Signalkomponente und eine mit dem zweiten Signal verkoppelte dritte Signalkomponente der ersten Kombinationseinrichtung zugeführt sind, deren Ausgang das erste modifizierte Signal liefert, und
• in der Modifikationsschaltung eine mit dem zweiten Signal verkoppelte vierte und fünfte Signalkomponente und eine mit dem ersten Signal verkoppelte sechste Signalkomponente der zweiten Kombinationseinrichtung zugeführt sind, deren Ausgang das zweite modifizierte Signal liefert.

The problem is solved as follows from the features of claim 1:

• Circuit for modifying a first and second signal from a signal source providing at least two signals, with means for forming signal components which are formed from the first and second signals and are combined into a modified first and modified second signal by means of first and second combination means, wherein
• in the modification circuit a first and second signal component coupled with the first signal and a third signal component coupled with the second signal are fed to the first combination device, the output of which supplies the first modified signal, and
• in the modification circuit a fourth and fifth signal component coupled to the second signal and a sixth signal component coupled to the first signal are fed to the second combination device, the output of which supplies the second modified signal.

• Fig. 1 zeigt eine erste bekannte Modifikationsschaltung,
• Fig. 2 zeigt eine zweite bekannte Modifikationsschaltung,
• Fig. 3 zeigte eine dritte bekannte Modifikationsschaltung,
• Fig. 4 zeigt ein erstes Ausführungsbeispiel der Erfindung und
• Fig. 5 zeigt ein weiteres Ausführungsbeispiel der Erfindung.

The invention and advantageous embodiments are now explained in more detail with reference to the figures in the drawing:

• 1 shows a first known modification circuit,
• 2 shows a second known modification circuit,
• 3 shows a third known modification circuit,
• Fig. 4 shows a first embodiment of the invention and
• 5 shows a further exemplary embodiment of the invention.

The known modification circuits from FIGS. 1 to 3 are found, for example, in the article cited in Elrad 1994, number 7, pages 76 to 81. Each circuit has an input for a left signal L and another input for a right signal R. Accordingly, each modification circuit has an output for a modified left signal L ' and another output has a modified right signal R ' . Furthermore, each circuit contains a first and second combination device K1, K2, in which different signal components are combined with one another, as a rule added or subtracted, in order finally to form the modified output signal L ' or R ' . These modified signals are then each fed to at least one loudspeaker (not shown), and these must not be too close together. It is known that a directional impression can only be recognized if the two signals L, R and L ' , R ' differ from one another. The greater the difference, the greater the ability to differentiate, which ultimately means that the sound sources subjectively located by the listener move apart, as it were. The individual modification circuits increase the differences in the individual signals to enlarge the stereo base and, on the other hand, reduce the common signal component. The common signal component is usually referred to as a mono signal and the difference as a difference signal. As is well known, these two components play an important role in the transmission of the stereo multiplex signal. The mono component in itself sounds good. However, the difference is not an actual audible signal and alone sounds very uncomfortable.

Filter circuits are included in all of the circuits of FIGS. 1 to 5. For audio signals, however, it can be assumed that as a rule the lower ones Frequency components up to a few 100 Hz are available as mono signals and the directional dependence only affects the frequency components above. This takes into account the fact that low frequencies cannot be resolved by the human ear in any direction. The individual signal components that influence the right and left signal are thus high-pass filtered signals, so that the filter circuits operating in the forward direction are implemented by high-pass filters HP. The respective contribution of the individual signal components to the modification is controlled by multipliers M and weighting factors, which can be negative, positive and the amount greater than 1. In reality, the weighting factors move in relatively narrow areas, because otherwise the effects produced create a false sound impression. In FIGS. 2 and 3 there is only one multiplier M, which effects the weighting by means of a supplied signal k. In Fig. 1 there are two multipliers M, both of which are driven by the weighting factor k.

• 1. Das ursprüngliche erste und zweite Signal L bzw. R sind einander entgegengesetzt gleich: $\text{R = - L}$
  
  . Das bedeutet, daß das Summensignal (= Monosignal (R + L)) den Wert Null hat und das Differenzsignal (L - R) seinen Maximalwert erreicht.
• 2. Das ursprüngliche erste und zweite Signal L bzw. R sind einander gleich: $\text{R = L}$
  
  . Das bedeutet, daß das Summensignal (= Monosignal (R + L)) seinen Maximalwert erreicht und das Differenzsignal ( L - R) den Wert Null aufweist.
• 3. Eines der ursprünglichen Signale L bzw. R hat den Wert Null: z.B. R = 0. Das bedeutet, daß das Summensignal (L + R) und das Differenzsignal (L - R) dem Betrag nach gleich sind. Das Summensignal (R + L) täuscht bei der Modifikation gegebenenfalls ein unspezifisches Signal (Monosignal) vor, das bei einer unzweckmäßigen Schaltung das Modifikationsergebnis stört. Mit diesen extrem einseitigen Signalen L und R = 0 läßt sich anschaulich darstellen, inwieweit Signalkomponenten bei der jeweiligen Modifikationsschaltung in den falschen Signalzweig eingekoppelt werden.

The considerations on which the individual circuits of FIGS. 1 to 3 are based are described below. The following frequency diagrams show how the modifications then affect the individual modified signals using the following characteristic signal contents:

• 1. The original first and second signals L and R are mutually the same: $\text{R = - L}$
  
  . This means that the sum signal (= mono signal (R + L)) has the value zero and the difference signal (L - R) reaches its maximum value.
• 2. The original first and second signals L and R are the same: $\text{R = L}$
  
  . This means that the sum signal (= mono signal (R + L)) reaches its maximum value and the difference signal (L - R) has the value zero.
• 3. One of the original signals L or R has the value zero: for example R = 0. This means that the sum signal (L + R) and the difference signal (L - R) are equal in amount. The sum signal (R + L) may fake an unspecific signal (mono signal) during the modification, which interferes with the result of the modification if the circuit is unsuitable. With these extremely one-sided signals L and R = 0, illustrate clearly to what extent signal components are coupled into the wrong signal branch in the respective modification circuit.

In any case, the aim is to keep the frequency response as straight as possible after the modification, because otherwise the sound will be distorted. The overall impression of volume should not be changed either.

werden dagegen bei höheren Frequenzen angehoben. Signale mit einem höherem Monoanteil hören sich somit dumpf an und Signale mit einem höheren Differenzanteil werden bei hohen Frequenzen unangenehm hervorgehoben.In the known circuit arrangement of FIG. 1, the directional impression is amplified by subtracting part of the first and second signals L and R, which is determined by the weighting factor k and a high-pass filter HP, from the other signal R and L, respectively becomes. The frequency diagram on the left shows that this modification is ideal if either only a first signal L or only a second signal R is present. The associated frequency response L ' with R = 0 runs flat in this case. Signals that have a higher sum component L + R (i.e. R is approximately equal to L) appear to be lowered at higher frequencies. Opposing signals $\text{R = - L}$

are raised at higher frequencies. Signals with a higher mono component thus sound muffled and signals with a higher differential component are highlighted unpleasantly at high frequencies.

The known circuit example of FIG. 2 reinforces the directional impression by subtracting signal components with the same signal component, that is to say signals with a high sum component L + R, from the first and second signals L and R, as a result of which the differences in the first and second signals become more apparent. The frequency diagram on the left shows that this is optimal for a signal mixture that has a significant sum of R + L and additionally a highlighted or greatly reduced signal source, e.g. R = 0. For audio signals, this corresponds to a one-sided sound source and a high proportion of mono signals.

In the known circuit arrangement of FIG. 3, a difference signal L-R is finally formed from the first and second signals L and R by means of a subtractor sb and a signal component is formed therefrom by means of a high-pass filter HP and a weighting stage M, which adds to the first signal L. and from the second Signal R is subtracted. The difference between the first and second signals is increased by adding and subtracting the difference value, so that the modified signals L ' , R ' have an amplified directional effect at the output and thus enlarge the stereo base.

einen idealen Frequenzgang aufweist. Für unterschiedliche Signale, also im Grenzfall gegensinnig gleiche Signale $\text{R = - L}$

, verläuft der Frequenzgang jedoch sehr ungünstig. Diese Signale werden bei höheren Frequenzen bis zu 10 dB angehoben und verfälschen daher den Klangeindruck.The adjacent frequency diagram shows that the circuit of FIG. 3 for unspecific signals or mono signals $\text{R = L}$

has an ideal frequency response. For different signals, in other words the same signals in opposite directions $\text{R = - L}$

, the frequency response is very unfavorable. These signals are raised up to 10 dB at higher frequencies and therefore distort the sound impression.

The invention teaches that a general circuit with which all variants can be realized can be accomplished by including further signal components in the respective modification, the influence of which is controlled by associated weighting factors. The individual signal components are also combined by means of combination devices, ie added or subtracted, in order finally to obtain a modified first and second signal L ' or R ' again.

4 shows a first exemplary embodiment of the modification circuit according to the invention, in which each modified signal is formed by three signal components. In order to carry out any modifications, it should be possible to change each signal component individually using a filter circuit and a weighting factor as required. The embodiment according to FIG. 4 already represents a simplification insofar as two signal components s2, s3 and s5, s6 are routed via a single filter circuit F2 and F4.

A signal source q supplies a first and a second signal L, R at its output. The signal source q is not determined in more detail, it can also represent, for example, a multiple signal source with parallel outputs, the first and second signals being attributed to adjacent signals. For the sake of clarity, the exemplary embodiments are limited to stereo signals, the first signal L corresponding to a left signal and the second signal R corresponding to a right signal. In these cases, the signal source q contains a decoder for stereo multiplex signals.

In the circuit according to FIG. 4, the first signal L is fed to an input of a first combination device K1 by means of a first filter F1 and a first weighting device with a multiplier M1 as the first signal component s1. The associated weighting factor g is supplied to the first multiplier M1 as a data value or corresponds to a fixed position shift. The first signal L is also fed to the input of a second filter F2 and forms, by means of a second weighting device, a sixth signal component s6 which is fed to a second combination device K2, at the output of which the second modified signal R 'can be tapped. The weighting in the second weighting device effects a second multiplier M2, the weighting input of which is supplied with a second weighting factor k. After the second weighting device, the signal is passed through a third weighting device and reaches the first combination device K1 as second signal component s2. The weighting in the third weighting device effects a third multiplier M3, the weighting input of which is supplied with a third weighting factor α. In parallel to the first, second and sixth signal components s1, s2, s6 which are formed from the first signal L, a fourth, fifth and third signal component s4, s5, s3 are formed from the second signal R. A third or fourth filter F3, F4 corresponds to the first or the second filter F1, F2. The first, second and third multipliers M1, M2, M3 correspond to a fourth, fifth and a sixth multiplier M4, M5, M6, to which the first, second and third weighting factors g, k and α are supplied. The third and sixth signal components s3, s6 are routed to a subtrahead input of the first and second combination devices K1, K2. The subtrahend inputs can be avoided if the associated weighting factors are changed in the sign.

FIG. 5 shows another exemplary embodiment of the invention which contains the circuit of FIG. 4 in a simplified form. In addition, the circuit contains regulating devices, b1, b2, r and control devices st for regulating and / or specifying the weighting factors. The embodiment of the circuit according to FIG. 5 is oriented even more towards the processing of audio signals than the more general circuit from FIG. 4. The signal source q provides the first and second signals L, R a left and right signal. The first and fourth signal components s1, s4 are neither filtered nor weighted but correspond directly to the first and second signals L and R. By means of a high-pass filter HP and the weighting by the second weighting factor k, the first signal L is used to form the sixth signal component s6, which is fed to the subtrahend input of the second combination device K2 . In the same way, the third signal component s3, which is fed to the subtrahend input of the first combination device K1, is formed from the second signal R by means of a high-pass filter HP and the same weighting factor k. The second weighting factor k is controlled by the control device st, which thus influences the level of the desired effect and thus the stereo base width. Since, as already stated, the directionality is only given in the middle and upper frequency range for conventional stereo signals, high-pass filters HP, whose cut-off frequency is greater than 300 Hz and typically, are used to form the second, third, fifth and sixth signal components s2, s3, s5, s6 is at 700 Hz. It can be seen from the frequency diagrams of FIGS. 1 to 3 that the range of the signal increases or decreases is changed via the cut-off frequency, which has an effect on the auditory impression in the case of mixed signals. The second and fifth signal components s2, s5 are formed from the high-pass filtered first and second signals, respectively, by changing the size of the signal by means of the third weighting factor α. The value of the third weighting factor α can now be used to set not only the properties of the known circuits from FIGS. 1 to 3, but also any intermediate stages, which enables optimum signal adaptation.

The frequency diagram m of FIG. 1 is set with a weighting factor α = 0, which provides an optimal modification for stereo signals in which one of the two components L, R has the value 0. With a weighting factor α, which is approximately between 0.4 and 0.5, a frequency response is set which corresponds to the frequency response of FIG. 2 and is optimal for mixed signals. If necessary, the third weighting factor α can also be negative in order to reduce the signal boost for unspecific signals in the upper frequency range. Finally, the frequency response of FIG. 3 is set with a weighting factor α = 1, which is ideal for pure mono signals or signals with a high mono signal component.

The third weighting factor α can be set in different ways. Either as a fixed value via the control device st - this is shown in FIG. 5 by a dashed connection. However, the third weighting factor α can also be controlled adaptively by the signal properties themselves, which are determined, for example, by means of a first evaluation device b1 from the left and right signals L, R. In the simplest case, the mono or difference signal component is determined via adders or subtractors. Individual frequency ranges can be treated separately or specially weighted by means of individual filters with which the physiological hearing sensitivity is simulated, for example. This corresponds to an adaptive control of the weighting factor α, which is shown schematically in FIG. 5 by the dashed line at the output of the first evaluation device b1.

If a corresponding evaluation is also carried out on the modified output signals L ' , R ' by means of a second evaluation device b2, the outputs of the first and second evaluation devices b1, b2 can be connected to a control device r, the output of which controls the level of the weighting factors. By comparing the input and output signals of the modification circuit, the control device r can be used, in particular, to ensure that the volume impression does not change during the modification, regardless of the respective effect control. If the control device r is to take into account the volume impression in the entire frequency range or in individual frequency ranges, then the first and second evaluation devices b1, b2 must determine, among other things, performance-related data from the signals at the input and output of the modification circuit. In the exemplary embodiment of FIG. 5, the output of the control device r controls the third weighting factor α.

Of course, a combination of the evaluation and control devices from FIG. 5 with the modification circuit from FIG. 4 is also conceivable. Here, the proportion of the first or second signal L, R can be controlled via the first weighting factor g. The first and third filters F1, F3 can be connected through or can be implemented by an all-pass. The time equalizations required for digital circuits are, as usual, not shown in the individual circuit examples. It is again pointed out that the invention and the associated exemplary embodiments are in no way limited to the processing of stereo signals, but that the adaptive effect control is advantageous for many other signals.

Claims (10)

Circuit for modifying a first and second signal (L or R) supplied by a signal source (q) with devices (F1 to F4, M1 to M6) for forming signal components (s1 to s6) formed from the first and second signal and are combined into a modified first and modified second signal (L 'or R') by means of a first and second combination device (K1 or K2),
characterized by the following features: - A first and second signal components (s1, s2) coupled to the first signal (L) and a third signal component (s3) coupled to the second signal (R) are fed to the first combination device (K1) whose output is the first modified signal (L ') returns, and - A fourth and fifth signal component (s4, s5) coupled to the second signal (R) and a sixth signal component (s6) coupled to the first signal (L) are fed to the second combination device (K2), the output of which is the second modified signal (R ') returns. Circuit according to Claim 1, characterized in that the devices for forming the first to sixth signal components (s1 to s6) contain filters (F1 to F4) and weighting devices (M1 to M6). Circuit according to claim 2, characterized by the following features: the first signal component (s1) is formed from the first signal (L) by means of a first filter (F1) and a first multiplier (M1), which effects a first weighting factor (g), the sixth signal component (s6) is formed from the first signal (L) by means of a second filter (F2) and a second multiplier (M2), which brings about a second weighting factor (k), the fourth signal component (s4) is formed from the second signal (R) by means of a third filter (F3) and a fourth multiplier (M4) which effects the first weighting factor (g), the third signal component (s3) is formed from the second signal (R) by means of a fourth filter (F4) and a fifth multiplier (M5) which effects the second weighting factor (k), the second signal component (s2) is formed from the sixth signal component (s6) by means of a third multiplier (M3) which effects a third weighting factor (α), - The fifth signal component (s5) is formed from the third signal component (s3) by means of a sixth multiplier (M6), which effects the third weighting factor (α) - The respective value of the weighting factors (g, k, α) is either predetermined by a control device (st) or dependent on evaluation devices (b1, b2). Circuit according to one of Claims 1 to 3, characterized in that at least one of the weighting factors (g, k, α) is controlled by means of a first evaluation device (b1) to which the first and second signals (L, R) are fed. Circuit according to one of claims 1 to 3, characterized in that a first evaluation device (b1) the first and second signal (L, R) and a second evaluation device (b2) the modified first and modified second signal (L ', R') are supplied and a control device (r), which is coupled to the first and second evaluation device (b1, b2), at least one of the Weighting factors (g, k, α) controls. Circuit according to one of Claims 1 to 5, characterized in that the signal source (q) is an audio signal source which delivers a left signal as the first signal (L) and a right signal as the second signal (R). Circuit according to Claim 6, characterized in that the second and fourth filters (F2, F4) are each a high-pass filter and that the respective stereo effect is changed by means of the second weighting factor (k). Circuit according to one of Claims 1 to 7, characterized in that the first signal component (s1) corresponds to the first signal (L) and the second signal component (R) corresponds to the second signal (R). Circuit according to Claim 5, characterized in that the control device (r) influences the value of the third weighting factor (α). Circuit according to Claim 5 or 9, characterized in that power levels in individual frequency ranges are determined in the first and second evaluation devices (b1, b2) and the same frequency ranges are processed in the control device (r).

Images