WO2014041385A2

WO2014041385A2 - A device simulating the distortion of human ear and a method to process a sound signal

Info

Publication number: WO2014041385A2
Application number: PCT/HU2013/000093
Authority: WO
Inventors: Viktor BAGDÁN; László CZIMERMAN; Kálmán MÁTHÉ; József PYTEL
Original assignee: Pécsi Tudományegyetem
Priority date: 2012-09-14
Filing date: 2013-09-13
Publication date: 2014-03-20
Also published as: WO2014041385A3; HU231071B1

Abstract

The invention relates to a device and a method to process audio signals. Said device comprising an input, an output and a continuous signal propagation path extending from the input to the output for audio signals applied to said input. The device further comprises in the signal propagation path from said input towards said output in the given order a first linear distortion generating unit (20) having first transfer characteristics, a nonlinear distortion generating unit (30) having nonlinear transfer characteristics to generate frequency-dependent distortion of the linearly distorted audio signal, and a second linear distortion generating unit (40) having second transfer characteristics, wherein said second transfer characteristics corresponds essentially to the inverse of said first transfer characteristics. The inventive method comprising subjecting said audio signals to first linear distortion characterized by first transfer characteristics, then subjecting the obtained linearly distorted audio signals to frequency-dependent nonlinear distortion, and then subjecting the thus obtained nonlinearly distorted audio signals to second linear distortion characterized by second transfer characteristics, said second transfer characteristics being provided essentially as the inverse of said first transfer characteristics.

Description

A DEVICE SIMULATING THE DISTORTION OF HUMAN EAR AND A METHOD TO PROCESS A

SOUND SIGNAL

The present invention falls into the field of high-quality sound amplification. In particular, it relates to an electronic device - especially to an audio amplifier - to process audio signals that gives perfectly realistic auditory experience even at lower sound pressure levels. The present invention also relates to a method of this kind to process audio signals, as well as to a sound amplification process based on said method.

Nowadays, devices produced at continuously decreasing costs and capable of producing high sound pressure levels (SPL) are rapidly spreading. Without being exhaustive, high-efficiency Class-D amplifiers of portable music playback equipments, as well as audio amplification devices integrated into audio amplifying equipments used as parts of sound systems installed in vehicles and/or flats or e.g. at various public events represent a few examples of such devices. The big- gest probable drawback of these devices roots in their capability to easily exceed the so-called equivalent sound pressure level of 85 dBA measured for eight hours that - according to medical examination - can result in psychic disturbances, in more severe cases in sleep disturbances and/or in extreme cases even in permanent hearing loss as well due to exposure to sound. In addition, studies performed among young people showed that due to the excessive sound pressure levels used to achieve the desired auditory experience, regular music listening with ear- buds and headphones or in concert halls results in statistically measurable/detectable hearing loss in this group of people.

It was also observed that the perception of sound spatialization (i.e.„sur- round sensation") gets worse in home theatre systems as the physical, measurable sound pressure level increases/is elevated. Thus, users generally have to make a compromise between surround sensation and the desired auditory experience when such systems of today are made use of.

Besides the unhealthy effects, environmental pollution caused by the de- vices in question cannot be neglected either: the desired loudness sensation achievable at the relatively high sound pressure levels generated by said devices represents rather significant power consumption. Furthermore, the noise emission to the environment by said devices is not desired either.

There exist processes by the application of which the loudness sensation can be enhanced without increasing further the sound pressure level, for example by exploiting the equal loudness contours (or phon curves), as it is discussed in relation to the procedure disclosed in U.S. Patent No. 5,172,358. Due to the formation of sound spectra differing from natural sound spectra, these techniques lead to unpleasant sound sensation and they, in general, spoil user auditory experience. Presently used methods to enhance loudness sensation apply compres- sors, as it is discussed e.g. in U.S. Patent No. 6,965,676.

In our studies we concluded that by setting a higher harmonic frequency spectrum similar to the distortion of human ear (or "imitating" it), an increase in the loudness sensation can be achieved over the entire audible frequency domain without degradation of sound quality. Accordingly, if a device for audio signal processing, in particular an audio amplifier, would be capable of producing this kind of higher harmonic range artificially via the combination of various electronic units included therein, the inclusion of said device in audio amplifying equipments would allow to avoid the usage of unnecessarily large and harmful sound pressure levels.

It should be here noted that due to the subjective nature of loudness sensation (and usually of hearing itself), the nature and frequency composition of said higher harmonic range (to be imitated) can be derived by statistical processing of measurements performed over a plurality of patients, for example. Moreover, comparing the available hearing test results of a certain patient with the statistical average would also facilitate the construction of an audio amplifier with a range of higher harmonics tailored to individual needs (hearing).

In light of the foregoing, the object of the present invention is to provide a device for audio signal processing, in particular an audio amplifier, as well as a method to process audio signals, in particular a sound amplification process, the development of which is based on generating artificially a range of higher harmonics similar to the one obtained due to human ear distortion. The invention also aims at facilitating to avoid the need for amplification at higher sound pressure levels than what is required to achieve the desired auditory experience, and thereby to alleviate or, alternatively, to eliminate health and/or environmental issues arising due to such excessive sound pressure levels.

The objective related to the realization of a device of this kind is achieved by developing a device for audio signal processing according to Claim 1. Possible preferred embodiments and applications of the device according to the present invention are defined by Claims 2 to 9. The objective related to the implementation of the above method to process audio signals is achieved by developing a method in accordance with Claim 10. Possible preferred further variants of said method to process audio signals, as well as its application to sound amplification are defined by Claims 11 to 13.

When practising said device and/or method for audio signal processing, known psycho-acoustic parameters characteristic of the loudness sensation of human ear are modelled (i.e. Jmitated") with high-precision by the combination, suitable setup of per se known electronic units. The essence of the present invention is the frequency dependent harmonics enrichment induced by said setup in a reliable and reproducible manner: the ratio of the even and the odd higher harmonics to the fundamental harmonic can be varied and/or is set as a function of the dynamics, signal level and/or frequency in a controlled manner by the device according to the invention. The set ratio corresponds basically to the range of higher harmonics appearing due to the distortion of human ear and can be determined by statistical means. It should be here noted that while the operation parameters of the thus obtained device can be precisely measured, set and reproduced, the sound spectrum generated on the output of said device and/or the mu- sical experience optionally created thereby cannot be empirically defined but can be statistically examined and characterized on the basis of high averages within studies performed with the participation of a plurality of patients.

In what follows, the invention will be discussed in more detail with reference to the drawings and in relation to a preferential application field (i.e. sound rein- forcing) of the invention, wherein

- Figure 1 shows a schematic block diagram of an audio amplifier based on a device for processing an audio signal according to the invention; - Figure 2 illustrates the connection scheme of a further embodiment of the audio amplifier based on the device for processing an audio signal according to the invention comprising a tube amplifier built with a double triode.

Figure 1 schematically illustrates an audio amplifier 100 constructed by means of utilizing a device for processing audio signals according to the present invention. In particular, an input sound signal (from now on, a signal) from a sound source 1 (e. g. a microphone or the output of a further sound system) is applied on an input IN of said amplifier 100, if needed by means of a suitable input means (e.g. an A/D converter, a signal conditioning unit, etc.). Then, said input signal passes to a sound-generating means 2 connected to an output OUT of said audio amplifier 100 via one or more stages realized in the form of one or more modules 10, 20, 30, 40, 50 in the present embodiment, wherein said stages are in signal- transmitting coupling with one another along a signal propagation path directed towards the output OUT of the audio amplifier 100. Said sound-generating means 2 can be equally provided as loud speaker, headphone, ear-buds or any other acoustic generator constructed in the form of a single unit or as a combination of several units.

In one embodiment, as it is shown in Figure 1 , the audio amplifier 100 is provided as a setup (or, in particular, a connection scheme corresponding to it), wherein said input IN is electrically connected to (or identical with) at least one input of module 10 having the functions of a pre-amplifier and a loudness controller, an input of module 20 functioning as a linear distorter is electrically connected to the output of said module 10, an input of module 30 functioning as a non-linear distorter is electrically connected to the output of said module 20, an input of mod- ule 40 functioning as an inverse linear distorter is electrically connected to the output of said module 30, an input of module 50 having the functions of a final amplifier and a loudness controller is electrically connected to the output of said module 40, and the output of said module 50 is connected to (or identical with) the output OUT of the amplifier 100. That is, said audio amplifier 00 is provided here as a chain of per se known electronic units connected electrically in series; the device for processing audio signals according to the present invention forms an integral part of this chain. It is also apparent to a person skilled in the art that said audio amplifier 100, as well as the device for processing audio signals itself, can be provided in the form of an ordered configuration of components that may optionally be capable of performing more than one of the above discussed functions.

Preferably module 10 is provided by a variable-gain (pre-)amplifier unit. The gain set for said module 10 (that is, the amplification factor of module 10) determines the operating range of the module 30. By an increase in the signal level, newer and newer higher harmonics are generated (that is, a harmonic enrichment takes place), and simultaneously cross modulation artefacts appear and begin to increase; in particular, sounds of difference frequencies are generated and sub- harmonics arise that is of significant importance especially in the ultrasonic domain.

Module 20 functioning as a linear distortion generating unit performs, in particular, pre-emphasis of the input signal, while module 40 functioning as an inverse linear distortion generating unit carries out, in particular, de-emphasis of the signal passing through it. Here, the intensity (signal level) of frequency components located at the (lower and upper) edges of the frequency range of the input signal is increased at pre-emphasis and is decreased at de-emphasis. The mid portion of the frequency range of said input signal is basically not affected by the modules 20 and 40. The width of the range of pre-emphasized/de-emphasized frequencies, as well as the transfer characteristics of said modules 20 and 40 are chosen to be in harmony (that is, in close approximation) with human ear distortion characteristics determined statistically. Modules 20 and 40 serve to correct the frequency dependence of loudness sensation, their operation is controlled by means of a concurrent control that is realized by exploiting suitable concurrent control elements. The concurrent control elements of modules 20 and 40 serve to adjust the cut-off frequencies, as well as other filtering parameters (the extent of pre-emphasis/de- emphasis), the frequency and phase responses of the transfer signal remain unchanged upon their influence.

Module 30 serves to practically generate nonlinear distortion which is the core of the inventive solutions. A frequency dependent distortion of the signal entering said module 30 is basically induced by two effects: (i) on the one hand, signals of pre-emphasized frequency with higher signal levels will fall on a portion of the nonlinear transfer characteristics of the module 30 that is characterized by a larger curvature, and

(ii) on the other hand, the negative feedback, which is higher than the negative feedback used for the pre-emphasized signals, used for the non pre-emphasized signals forming a mid portion of the frequency range of the input signal lowers the distortion of harmonics over the mid portion of the frequency range which results in the enhancement of the auditory experience (and thus e.g. of comprehensibility).

By altering the nonlinear transfer characteristics of the module 30, the ratio of the even and the odd higher harmonics to the fundamental harmonic is modified and set. Said setting is performed particularly in such a way that this ratio approximate, as close as possible, a similar ratio corresponding to human ear distortion that has been previously determined via statistically analyzing test results obtained in studies performed with the participation of a great number of human be- ings.

As it is apparent to a person skilled in the art, said module 30 with the above function can be provided in the form of various electronic units; it can be realized e.g. by a (double) triode differential amplifier, a pentode amplifier, as well as other similar units. Depending on the practical realization of the module 30, said setting of the ratio of the even and the odd higher harmonics to the fundamental harmonic can be done by altering the symmetry of the triode differential amplifier, by suitably regulating the auxiliary grid of said pentode amplifier, and optionally by means of analogue and digital multiplier circuits or via any other devices with nonlinear transfer characteristics.

Preferably, module 50 is also provided by a variable-gain (final) amplifier unit. Said module 50 serves to correct the amplification level set in module 10; to maintain a constant overall signal level, if an increasing pre-amplification is applied in module 10, module 50 will effect a decreasing final amplification that is in harmony with the increasing pre-amplification. That is, the signal level control of said modules 10 and 50 is of opposite sense, i.e. the product of the amplification factors of the modules 10 and 50 is basically constant; this ensures setting of the dis- tortion over the entire dynamic and frequency ranges and thereby a basis for the generation of the higher harmonics required.

The above mentioned concurrent (that is, synchronized) control is performed by a control unit 60 that forms part of said audio amplifier 100 and is con- nected with each module 10, 20, 30, 40, 50 according to needs. Said control unit 60 comprises mechanical or electronic control elements. Said control elements are provided by e.g. potentiometers mounted onto a common shaft, disc-type switches or, alternatively, electronically controllable multiplier circuits. The control unit 60 serves for fine-tuning and tailoring the audio device 100 to individuals' ears.

Figure 2 illustrates a connection scheme for a possible further exemplary audio amplifier 200 practically realized also by utilizing the device for processing audio signals according to the invention. The modules 10, 20, 30 of Figure 1 being functionally separated from each another are combined now into a first stage 210, while modules 40, 50 of Figure 1 are combined into a second stage 220, wherein said first and second stages 210, 220 are mutually electrically connected. The input signal is applied to an input IN of the first stage 210, then it propagates through the stages 210, 220 and is obtained back on an output OUT of the second stage 220.

Pre-amplification of the input signal fed into the first stage 210 is performed by an operational amplifier IC1A and its auxiliary circuits. Linear distortion of the thus obtained signal is effected by the elements R7-C2 and R9-C10 in the negative feedback. Then, setting of the desired ratio of the even and the odd higher harmonics to the fundamental harmonic is executed by a double triode 212 in cooperation with a potentiometer R15. Said first stage 210 creates a negative feed- back at the intermediate frequencies that is greater than the one created thereby at the edges of the frequency range; the frequency domain of the signal passed through the first stage 210 exhibits pre-emphasis at its lower and upper edges, while - compared to the input signal - it is much more linear in its mid portion. The transfer characteristics of the double triode 212 can be modified by further resis- tors and capacitors in said first stage 210, as it is obvious to a person skilled in the art. Then, the signal modified in the first stage 210 in accordance with a characteristics that is in relatively good approximation (practically is identical) with human ear distortion gets into the second stage 220, wherein setting the extent of the inverse linear distortion, in particular of de-emphasis takes place by means of a variable resistor R14. After this, the signal reaches its final intensity in an operational amplifier IC1 B and then appears on the output OUT of the second stage 220, to which output preferably a sound-generating means can be connected.

It is apparent to a person skilled in the art that the modular setups or connection schemes constituting the devices according to the present invention can be equally accomplished by analogue techniques based on traditional analogue units and via exploiting digital circuit components by means of e.g. computers and/or digital technology using specific digital signal processing processor(s) (DSP).

By means of the sound amplification process accomplished by using a device for processing audio signals, the frequency spectrum and/or the dynamics spectrum of an audio signal received from an audio source are/is (actively) modi- fied so as to produce an audio signal of modified frequency and/or dynamics characteristics for a listener. In the modification, the ratio of the even and the odd higher harmonics to the fundamental harmonic is altered/set in a controlled way as a function of signal level and/or frequency so as to closely approximate a characteristics corresponding to human ear distortion and derivable by statistical means. In this way it is achieved that listener's auditory experience will be excellent even at relatively low sound pressure levels. Or putting this another way, by exploiting the present inventive solutions, the same acoustical experience can be achieved at lower sound pressure levels.

By means of the audio amplifier/sound amplification process that can be ob- tained through utilizing the device/method for processing audio signals according to the present invention, loudness sensation of human ear can be significantly enhanced without changing the auditory experience, that is, from practical point of view, the tonality of sound having passed through the device; the obtained audio amplifier provides a totally linear frequency response over its entire dynamic range. Furthermore, upon psycho-acoustical considerations it can be stated that sound amplification achieved by the present audio amplifier influences the ratio of sound pressure to sound sensation rather favourably and also enlarges the dy- namic range perceived by human ear. In practice, the audio amplifier discussed above ensures this by providing a transfer characteristics that simulates/imitates the distortion characteristics of human ear over a rather broad frequency range.

By the application of the inventive solutions, the sound pressure level can be significantly decreased without deteriorating musical experience, and thus various psychic disturbances/diseases and/or hearing impairment can be avoided/prevented. The extent of noise pollution by the audio amplifying equipments installed at various public events or in flats can also be significantly decreased by means of the present inventive solutions. Furthermore, the extent of environmental pollution can be decreased as well; we found that the power requirement of an audio amplifier built with utilizing a device for processing audio signals according to the invention can be, preferably, at least five times, but even ten times smaller that that of the audio amplifiers available nowadays.

The device for processing audio signals according to the invention can be matched (as a separate unit, for example) with any kind of sound system available nowadays or can be provided as an integrated part of such systems. Preferably, the device according to the invention can be used in ear-buds or headphones or as a (n inner or outer) part of other mobile equipments meeting high musical demands. Furthermore, it can be used in studio equipments, as well as e.g. in guitar amplifiers, in particular. Its application in vehicles can be considered especially preferred due to the relatively small acoustical room available there. Moreover, the device according to the invention can be equally used in case of multichannel home theatre systems to improve surround sensation.

Moreover, the device for processing audio signals according to the inven- tion can be tailored to individual needs as well via performing audiometry on a targeted patient, analyzing his/her hearing and then comparing the test result(s) with stored (e.g. in the device itself) parameters characteristic of hearing that have been obtained statistically. That is, said device can be adjusted in such a way that the targeted patient receive the desired auditory experience with emitting the least possible sound pressure therefor.

Claims

1. A device to process audio signals comprising an input (IN), an output (OUT) and a continuous signal propagation path extending from the input to the output for audio signals applied to said input, characterized in that it comprises in the signal propagation path from said input towards said output in the order given below

- a first linear distortion generating unit (20) having first transfer characteristics,

- a nonlinear distortion generating unit (30) having nonlinear transfer characteristics to generate frequency-dependent distortion of the linearly distorted audio sig- nal, and

- a second linear distortion generating unit (40) having second transfer characteristics, wherein

said second transfer characteristics corresponds essentially to the inverse of said first transfer characteristics.

2. The device according to Claim 1 , characterized in that said first transfer characteristics corresponds to a statistically derived distortion characteristics of human ear.

3. The device according to Claim 1 or 2, characterized in that said nonlinear transfer characteristics of said nonlinear distortion generating unit (30) corre- sponds to a transfer characteristics that reproduces essentially a ratio of the even and the odd higher harmonics to the fundamental harmonic, wherein said ratio is deduced from the statistically derived distortion characteristics of human ear.

4. The device according to any of Claims 1 to 3, characterized in that said nonlinear distortion generating unit (30) is chosen from the group comprised of tri- ode differential amplifiers, pentode amplifiers and digital signal processing processors.

5. The device according to any of Claims 1 to 4, characterized in that it further comprises

- a first amplifier unit (10) to amplify the signal level of the audio signal by a de- sired first amplification factor, and - a second amplifier unit (50) to amplify the signal level of the audio signal by a desired second amplification factor, wherein

output of said first amplifier unit is connected to the input of the signal propagation path,

input of said second amplifier unit is connected to the output of the signal propagation path, and

said first and second amplifier units are provided with a control unit (60) for synchronized control of said amplifiers, said control unit (60) for synchronized control being configured to vary said first amplification factor and said second amplifi- cation factor in opposite sense.

6. The device according to Claim 5, characterized in that said control unit for synchronized control of the first and second amplifier units is configured to maintain the frequency response and the phase response of the fed audio signal essentially unchanged.

7. The device according to Claim 5 or 6, characterized in that said control unit for synchronized control is chosen from the group comprised of potentiometers mounted onto a common shaft, disc-type switches and electronically controllable multiplier circuits.

8. An audio amplifier ( 00; 200), characterized in that it comprises a device to process audio signals according to any of Claims 1 to 7.

9. A sound system, characterized in that it comprises a device to process audio signals according to any of Claims 1 to 7 and/or an audio amplifier (100; 200) according to Claim 8, or it is configured to be capable of electrically connected to at least one of said device and said audio amplifier.

10. A method to process audio signals, comprising subjecting said audio signals to first linear distortion characterized by first transfer characteristics, then subjecting the obtained linearly distorted audio signals to frequency-dependent nonlinear distortion, and then subjecting the thus obtained nonlinearly distorted audio signals to second linear distortion characterized by second transfer charac- teristics, said second transfer characteristics being provided essentially as the inverse of said first transfer characteristics.

11. The method to process audio signals according to Claim 10, comprising generating said first transfer characteristics corresponding to a statistically derived distortion characteristics of human ear.

12. The method to process audio signals according to Claim 10 or 11 , comprising performing said nonlinear distortion in accordance with a transfer characteristics reproducing essentially a ratio of the even and the odd higher harmonics to the fundamental harmonic, said ratio being deduced from the statistically de- rived distortion characteristics of human ear.

13. A sound amplification process, comprising subjecting audio signals to first amplification characterized by first amplification factor before the audio signals being processed by the method to process audio signals according to any of Claims 10 to 12, and subjecting the audio signals having been processed by the method according to any of Claims 10 to 12 to second amplification characterized by second amplification factor, further comprising maintaining the product of said first amplification factor and said second amplification factor essentially at a constant value during performing said first and second amplifications of said audio signals.