EP3695620B1 - Improved sound transducer - Google Patents

Description

The present document relates to a device, a method, a signal processing unit, data for acoustic reproduction, a sound transducer, in particular headphones or earphones, and a software product for improving sound reproduction.

From the pamphlet U.S. 2007/154049 A1 a device according to the preamble of claim 1 is known. Further prior art can be found in the publication WO 2014/138735 A1 and the print U.S. 2008/107287 A1 be removed.

Problems with the reproduction of sound signals via headphones are known from the prior art, so that when sound events are emitted via headphones, these sound events are perceived by the human ear significantly differently under certain conditions than with sound sources that are distant from the ear, such as loudspeakers. Despite the use of transfer functions in the outer ear, spatial imaging errors (elevation angle) can occur if the sound source is in the median plane (imaginary plane perpendicular between the ears) of the listener. With such correlated signals, interaural levels and transit time differences are then missing. In the case of sound sources in front, in particular, the sound signals are often perceived in the head or very close to the head (so-called in-head localization). The ICS often occurs in connection with a disturbing elevation (location at the top of the head). So far, these problems can only be improved by technically complex optical support or by head tracking. Other aberrations relate to the perceived volume of sound signals emitted via headphones. Compared to distant sound sources, headphones may be perceived as quieter, even though the sound pressure level is the same. It has been shown that this so-called SLD effect (Sound Pressure Loudness Divergence) always occurs together with the localization in the head.

The present invention is therefore based on the object of eliminating or at least reducing the above problems in order to achieve improved sound reproduction.

The object set here is achieved according to the invention by the subject matter of claim 1 .

In the following, a non-claimed device for acoustic reproduction is proposed, wherein the device is provided with a first electro-acoustic sound transducer for generating a sound field and wherein the first electro-acoustic sound transducer has an input for receiving an electrical signal for generating the corresponding sound field, the device thereby characterized in that a device is also provided which is set up to enter into an acoustic interaction with the sound field generated by the first electroacoustic sound transducer in order to generate a modified To generate a sound field and it is provided that the modified sound field has a predetermined acoustic impedance value.

The device can be at least one acoustic resonator and/or at least one further electroacoustic sound transducer. An electroacoustic sound transducer is therefore proposed either in cooperation with at least one further electroacoustic sound transducer or in cooperation with at least one resonator. In this case, for both of the aforementioned variants, it applies that an acoustic interaction is aimed at generating a modified sound field, so that the modified sound field has a predetermined acoustic impedance value. Alternatively, it is provided that both of the aforementioned variants are set up to set different impedance values or variable impedance values for the modified sound field.

The first electroacoustic sound transducer and/or the further electroacoustic sound transducer can be set up to receive an electrical signal as a function of impedance information and to convert it into an acoustic signal so that the modified sound field has a predetermined acoustic impedance value as a result of the corresponding acoustic interaction.

The acoustic resonator can be designed as a recess, a hole or as a Helmholtz resonator, with this being implemented in particular on the housing of the device, in particular in the inner and/or outer housing area.

The first electroacoustic sound converter and/or the further electroacoustic sound converter and/or the acoustic resonator can be controllable by means of a corresponding electrical signal in order to set different acoustic impedance values in the modified sound field. In this case, the control can take place either directly via the electrical audio signal to be fed in and/or via a separate signalling.

Furthermore, one of the devices of the above type is proposed, the device having a measuring unit, in particular a microphone for measuring a sound field parameter in order to be able to derive a given impedance value in the sound field in order to enable a subsequent electrical adaptation signal to be generated. It is expressly pointed out here that one or more of the embodiments proposed according to the invention are set up to actively carry out a measurement via a control loop in order to measure a current impedance value in the sound field in order to subsequently implement readjustment by generating a suitable signal.

The device is designed as a headphone or earphone. In particular, a corresponding housing for accommodating the device can be provided and designed as a helmet.

Furthermore, a device is proposed in which the position and/or the alignment of the first electroacoustic sound transducer and/or the further electroacoustic sound transducer and/or the acoustic resonator is designed to be changeable and can be changed and adjusted as required, in particular by means of a suitable electrical signal. This means in particular a variability in the position and/or orientation of a sound transducer or resonator. Furthermore, in the case of a resonator, the frequency response and/or the oscillating mass can be designed to be controllable.

According to a further aspect, a signal processing unit for processing signals for acoustic reproduction is proposed, which is set up to process a further signal for acoustic interaction with the first sound field as a function of a first signal which is provided for generating a first sound field in order to to generate a modified sound field, it being provided that the modified sound field has a predetermined acoustic impedance value.

Furthermore, a signal processing unit is proposed, the signal processing unit providing a factor depending on at least one sound pressure signal and/or a sound velocity of the first signal in order to generate a modified sound field, it being provided that the modified sound field has a predetermined acoustic impedance value.

Furthermore, a signal processing unit is proposed, the sound pressure signal and/or the sound velocity being derived by means of a measurement, in particular by means of at least one microphone.

Furthermore, a signal processing unit is proposed, the signal processing unit being set up to process an impedance signal so that the impedance signal can be provided to a sound transducer.

Furthermore, a signal processing unit is proposed, the modified sound field having a temporally predetermined variable acoustic impedance value.

A signal processing unit is also proposed, the signal processing unit being set up to process further relevant acoustic parameters, in particular geometric parameters of a headphone or an earphone, in order to set the predetermined acoustic impedance value for the modified sound field. Furthermore, data for acoustic reproduction are proposed which have data elements for acoustic interaction with a first sound field, the data elements being set up to generate a modified sound field, it being provided that the modified sound field has a predetermined acoustic impedance value.

Furthermore, data are proposed, the data elements being set up and designed in such a way that they can be converted into a corresponding electrical signal in order to be reproduced in a later step by an acoustic resonator and/or by at least one electroacoustic sound transducer.

The data elements can have impedance information.

Furthermore, data are proposed, the data containing control data for controlling the acoustic resonator and/or the at least one electro-acoustic sound transducer.

Furthermore, data are proposed, with the data being generated using one of the above signal processing units.

Furthermore, a processing device for processing and/or reproducing the data is proposed, the data corresponding to one of the above data variants and the processing device being in particular a smartphone, notebook, laptop, tablet PC, personal computer, wireless transmitter or server.

Furthermore, a sound converter is proposed, the sound converter being set up for playing back a generated signal by means of one of the signal processing units proposed above and/or data proposed from above.

Furthermore, a software product, which is stored on a storage medium and can be processed by an electronic data processing unit, is adapted to implement one of the signal processing units presented above and/or to generate or reproduce data presented above. Furthermore, a method for acoustic reproduction is proposed, the method having the following steps: generating a first sound field and generating a second signal for acoustic interaction with the first sound field in order to generate a modified sound field, it being provided that the modified sound field has a predetermined has acoustic impedance value.

It is therefore provided that, in addition to the sound pressure, information about the sound field impedance at the ear entrance is also taken into account in order to obtain reliable spectral information on the sound source localization even with correlated signals from the median plane receive. However, the hearing can only carry out such impedance information from the position of the eardrum at the end of the auditory canal.

The invention is characterized in particular by the fact that a headphone or earphone as part of a system according to the invention not only simulates the sound pressure signal, but also the sound field impedance generated by a distant sound source at the ear in order to improve negative phenomena such as IKL or SLD or entirely to avoid. In contrast to current binaural technology, the headphones ideally do not receive a sound pressure signal that contains head-related sound pressure frequency responses, since these are self-adjusting if the sound field impedance is set correctly in the headphones. The so-called outer ear transfer function (HRTF) only describes the relationship between the two ears. In the following, the procedure is described as to how the sound field impedance considered relevant for hearing is defined and how this can be measured.

Furthermore, a method for measuring head-related sound field impedances of headphones is proposed.

For the development of a headphone, a measurement method is required that indicates whether a headphone generates a sound field impedance that is relevant in terms of avoiding IKL and SLD. This is indicated when the measurement method with headphones provides the same result as with loudspeakers. The proposed measurement method expands the known method for determining the head-related sound pressure transfer function (outer ear transfer function, HRTF) by a second transfer function that contains information about the sound field impedance. With the help of a suitable artificial head at the end of each ear canal there is a so-called impedance microphone, which is able to deliver both a pressure signal and a speed signal from a power source (see below), a measurement test bench can be set up that can be used both for Loudspeaker as well as headphone sound reinforcement is suitable to determine a signal S _p dependent on the sound pressure and a signal Sz dependent on the sound pressure and the sound field impedance ( figure 1 ).

When the artificial head is exposed to loudspeakers by the signal S, the signals S _p and Sz result at the outputs of the microphones for the left and right ear. These signals depend on the frequency and the angle of incidence of the sound. If the artificial head is now exposed to the same signals via headphones (with possible signal processing), the signals S' _p and S' _Z are measured.

For headphones, which are also supplied with the signal S _p and which simulate sound field conditions comparable to a loudspeaker, the following applies: $$\mathrm{S}{\prime }_{\mathrm{p}}=\mathrm{k}{\mathrm{S}}_{\mathrm{p}}\mathrm{and}\mathrm{S}{\prime }_{\mathrm{Z}}=\mathrm{k}{\mathrm{S}}_{\mathrm{Z}}.$$

It should be noted here that the test bench for the headphones does not have to be an artificial head. A comparable measurement method, which is however limited to the sound pressure, has long been used in binaural technology to generate spatially perceptible sound fields in headphones. A sound pressure transfer function H _p can be determined from the measured signal S _p that no longer contains the loudspeaker frequency response by relating the head-related signals to the pressure signals of a free-field measurement without a head: $${\mathrm{H}}_{\mathrm{p}}={\mathrm{p}}_{\mathrm{Ear}}/{\mathrm{p}}_{\mathrm{free\; field}}$$

H _p describes the change in sound pressure caused by the presence of a human head (body) and the relationship between the ears. However, this function, also known as the outer ear transfer function (or HRTF) in current binaural technology, must be corrected for the sound pressure generated by this field impedance itself in a new headphone that simulates the sound field impedance. In the ideal case, H _p then only contains interaural relationships.

The signal Sz is new and, compared to the pure sound pressure signal S _{p ,} provides additional information about the sound field in front of the ear. It describes the acoustic resistance at the ear entrance of a human head, which a force source Q located in the ear canal feels when it exerts a force F _Q against an external sound field. The force F _Q is derived from the pressure in the ear canal by means of a suitable mechanism (a microphone that is not described in detail) and acts back on the sound field in phase with the pressure. For this reason the signal Sz is also dependent on the sound pressure. The force source Q is itself exposed to the force F _F of the external sound field. The force source Q thus impresses a force D _FQ =F _Q -F _F into the sound field and reacts with the velocity v _Q to the sound field impedance Z _F . Thus v _Q is generally a function of the sound pressure p and the sound field impedance Z _F : $${\mathrm{v}}_{\mathrm{Q}}=\mathrm{f}\left(\mathrm{p},{\mathrm{Z}}_{\mathrm{f}}\right)$$

Similar to the head-related sound pressure transfer function H _{p ,} an impedance transfer function Hz can be determined from the signal v _Q by relating v _Q to the signal of a free-field measurement without a head: $${\mathrm{H}}_{\mathrm{Z}}={\mathrm{v}}_{\mathrm{Q}-\mathrm{Ear}}/{\mathrm{v}}_{\mathrm{Q}-\mathrm{free\; field}}$$

Hz thus represents an extension of the previous head-related properties and can be used to characterize the properties of headphones with regard to the acoustic sound field impedance in front of the ear.

The application of the method described for measuring the signal Sz combined with a pressure sensor is referred to here as an impedance microphone. It is able to deliver both a sound pressure signal and a signal dependent on the sound field impedance.

With the help of the 2-microphone method, sound field impedance measurements are carried out on the outer ear of a test subject in order to characterize the differences in sound exposure with headphones and loudspeakers. Connections to the subjective hearing sensations IKL and SLD are also examined. It turned out that a measurement of the X-component of the sound field impedance depicts the differences quite well and gives an idea of the size and the frequency and angle dependence of the sound field impedance.

These impedance measurements are not identical to those made from the ear canal using impedance microphones and the method described above. They only apply to one component of the sound field in front of the ear.

Further following methods for influencing the sound field impedance in front of the ear in a headphone or earphone are presented.

The sound field conditions in front of the ear of a human head when exposed to sound from a distant sound source are modeled for a headphone or earphone that is characterized by an improvement in the localization in the median plane, in particular with regard to the forward localization. Ideally, a head-related impedance signal and a frequency-independent sound pressure signal are transmitted to the headphones. A vibration transducer impresses a proportional velocity signal into the headphone chamber and generates the corresponding head-related sound pressure at the specified sound field impedance. Alternatively, simplified systems can also be useful, in which the most important properties of the real sound field impedance at the ear are transferred to headphones.

1. a) die Schallfeldimpedanz vor dem Ohr soll einen überwiegend positiven Blindwiderstand im Frequenzbereich von ca. 100 Hz bis 2.5 kHz erhalten und/oder
2. b) bei Beschallung durch eine entfernte Schallquelle aus der Vorne-Richtung entstehen am Ohr zwei typische Schalldruckminima. In der Regel befinden sie sich in schmalen Frequenzbereichen um ca. 1 kHz und ca.2.5 kHz, je nach Kopf- und Körpergeometrie. Diese entstehen durch Minima in der Schallfeldimpedanz als Folge von Interferenzen. Diese Schalldruckminima werden erfindungsgemäß an den Kopfhörer nicht als Schalldrucksignal übergeben, vielmehr muss der Kopfhörer die entsprechende Schallfeldimpedanz annehmen, sodass als Folge davon diese Schalldruckminima entstehen und/oder
3. c) zur Realisierung von Richtungshören in der gesamten Medianebene werden die Minima in der Schallfeldimpedanz mit zunehmendem Schalleinfallswinkel zu tiefen Frequenzen verschoben, und zwar entsprechend dem Vorbild wie dies am Kopf durch Beschallung mit einer entfernten Schallquelle geschieht. Bei Schalleinfall von hinten sind die Minima in der Schallfeldimpedanz stark gedämpft oder verschwinden vollständig und/oder
4. d) erfindungsgemäß wird insbesondere eine Kalibriermöglichkeit am Kopfhörer realisiert, um individuelle Unterschiede von Hörern optimal ausgleichen zu können. Das kann sowohl die Größenordnung der Schallfeldimpedanz als auch die Lage der charakteristischen Minima beinhalten.

An embodiment according to the invention with modeling that approximates reality is characterized by one or more of the following properties:

1. a) the sound field impedance in front of the ear should have a predominantly positive reactance in the frequency range from approx. 100 Hz to 2.5 kHz and/or
2. b) when exposed to sound from a distant sound source from the front direction, two typical sound pressure minima occur at the ear. They are usually found in narrow frequency ranges around 1 kHz and 2.5 kHz, depending on head and body geometry. These arise from minima in the sound field impedance as a result of interference. According to the invention, these sound pressure minima are not sent to the headphones as a sound pressure signal handed over, rather the headphones must accept the corresponding sound field impedance, so that as a result these sound pressure minima arise and/or
3. c) In order to achieve directional hearing in the entire median plane, the minima in the sound field impedance are shifted to low frequencies as the angle of incidence of the sound increases, in accordance with the example of how this happens on the head when exposed to sound from a distant sound source. In the case of sound incidence from behind, the minima in the sound field impedance are strongly damped or disappear completely and/or
4. d) according to the invention, in particular, a calibration option is implemented on the headphones in order to be able to optimally compensate for individual differences between listeners. This can include both the order of magnitude of the sound field impedance and the position of the characteristic minima.

Devices, methods and processes that are able to influence the sound field impedance of a headphone are presented below.

According to figure 2 a modeling of the sound field impedance is realized with the help of sound transducer pairs. For this purpose, in addition to the sound pressure at the ear, a specific sound field impedance is achieved through the use of two sound transducers in a headphone capsule. With suitable signal processing, the desired sound field impedance can be influenced in the arranged direction. For this purpose, the sound pressures p ₁ , p ₂ and the sound velocities v ₁ , v ₂ of the individual sound transducers are first determined using a suitable impedance measurement method (2-microphone method) or by previously determined sound pressures and sound velocities based on geometry-related values. A factor k _F can then be calculated from this, which describes the signal difference between the two sound transducers. Multiple directions can be influenced by arranging additional pairs of speakers in other directions. FIG. 8 shows a simple principle with signal processing that calculates the signal K ₂ for the second loudspeaker as a function of the value of the sound field impedance present at the input. Signal conditioning can also be part of a computer simulation when Z _Fx changes over time, as with moving sound sources or when using head trackers. p ₁ , p ₂ and v ₁ , v ₂ are determined from individual measurements of the sound transducers Lsp1,2 using the 2-microphone method. S _p is the sound pressure signal and Z _Fx is the impedance information.

According to figure 3 a modeling of the sound field impedance with passive acoustic resonators is proposed. With the help of Helmholz resonators, the sound field impedance in headphones can be changed to positive reactances. The resonator consists of a tube with any cross-sectional area, one opening of which is in the volume between the ear and the sound transducer protrudes. Other resonators can also be used here. The accelerated air in the tube represents a mass, which together with the rigidity of the air volume forms a resonance system. The mass character of the sound field occurs above the resonance frequency. The bandwidth and the quality of the system can also be influenced with a flow resistance. A combination of several resonators can also be implemented.

According to an embodiment of the invention and with reference to FIG figure 4 a modeling of the sound field impedance with active electroacoustic systems is proposed. Acoustic impedances can be specifically modeled with a system consisting of a microphone, sound transducer, amplifier and a simulation function. Simple analog realizable examples are masses, springs, flow resistances or resonators. Digital networks are considerably more versatile, but require very low latency times. The principle is based on modeling the relationship between pressure and velocity in the KH pressure chamber. The pressure-signal proportionality of the microphone M and the signal-diaphragm velocity proportionality of the converter W _Z are important for the correct function. A simulation function describes the reciprocal of the desired acoustic impedance Z _F in the form of a transfer function Ua/Ue=v/p=l/Z _F . The emulation function reacts to the pressure signal at the input with a speed signal at the output. This signal controls the transducer W _Z , whose membrane executes a proportional speed. If the amount of air moved is large enough, it determines the sound field in the headphones. The replica can also have another input that can be used to control the shape of this transfer function. Below and with reference to figure 5 a function for analog simulation of sound field impedances in headphones is shown. The sound field at the ear of the human head in the free sound field and at low frequencies can be described in a first approximation as a plane wave and a scattered wave, which is reflected by a sphere considered to be "breathing".

A sound pressure is generated at the radiation impedance of the "breathing" sphere, which is superimposed on the plane wave. The resulting sound field impedance Z _F is equal to: $${\mathrm{Z}}_{\mathrm{f}}=\frac{\mathrm{p}}{\mathrm{v}}={\mathrm{Z}}_0\frac{2+\frac{1}{{\mathrm{fk}}_0\mathrm{a}}}{1+\frac{1}{{\mathrm{fk}}_0\mathrm{a}}}$$

The following example shows what an analog replica of 1/Z _F can look like. The example shows an additional simulation 2 of an interference that leads to a minimum in the sound pressure.

Continue as per figure 6 an earphone is suggested.

For earphones, in particular, the active electro-acoustic systems to influence the Sound field impedance interesting. The shape of the earphones is important here, as two sound transducers and a microphone can be accommodated in such a space-saving manner that they can still be worn comfortably by the listener. Further in figure 6 various arrangements of the sound transducers in an earphone are specified.

Claims (2)

1. System for acoustic reproduction, comprising:

   headphones or earphones having a microphone (M) for measuring a sound field parameter, a first electroacoustic sound converter for generating a sound field and a second electroacoustic sound converter which is configured to enter upon acoustic interaction with the generated sound field of the first electroacoustic sound converter in order to generate a modified sound field; and

   a signal processing unit, wherein the signal processing unit is configured to control the first electroacoustic sound converter and the second electroacoustic sound converter by means of electrical signals;

   characterized in that

   the signal processing unit is furthermore configured to determine from the measured sound field parameter a given impedance value in the sound field and to derive therefrom an electrical adaptation signal and to control the second electroacoustic sound converter on the basis of the electrical adaptation signal in order set acoustic impedance values in the modified sound field according to an impedance signal (S_Z).
2. System for acoustic reproduction according to claim 1, wherein the position and/or the alignment of the first electroacoustic sound converter and/or of the second electroacoustic sound converter can be altered and, where required, set by means of a suitable electrical signal.

Images