DE3922118A1 - Direction variable ear adapting for stereo audio transmission - involves outer ear transmission function tuning for binaural adapting - Google Patents

Abstract

The individual adapting of a stero audio transmission to the outer ear is carried out by individual binaural adapting of various local sound sources of respective headphone audio transmission systems. A receiver side directional signal is added for a reproduction side outer ear transmission function tuning for realising a natural, multidimensional locatisation. For the transmission of the reception- and reproduction-side directional signal pref. PCM is used. At the reception side may be used a suitable directional mixer console, simulating the linear, binaural interference and group transit times. Analog stereo signals are typically converted into PCM data. ADVANTAGE - Improved stereo reproduction.

Description

The invention relates to a method for full directionally variable outer ear customization in head-related stereo sound transmission according to the preamble of claim 1.

The specified sound transmission process should using a parallel to the sound recording created directional information codes, the in sound reproduction as a control variable a variably programmable or controllable Digital filter serves an individual directional, spatial and thus natural sound field reproduction via headphones enable.

It is known, to meet these needs, Headphones free field or diffuse field and / or after the outer ear transfer function or equalizing in a directionally neutral manner (K. Genuit: "Why Free field, notification of the radio exhibition Berlin 1983; Broadcast communications: Issue 1/1983, pages 17 to 26; Patent application: No. P 31 31 347.7-35; Progress in acoustics - DAGA 1987, pages 477 to 480).

The process is based on a replica the statistically sufficiently averaged Directional characteristic of the human ear, i.e. taking into account upper body, Hull, head and outer ear relief, which in the Realization or equalization of one Dummy head, directional mixer and headphones, be taken into account in a standardized manner.  

There are primarily two today major types of artificial head developments, which equalized in free field and diffuse field Structure versions (Brüel & Kjaer catalog: "Head and Body Simulator 4128"; Patent registration: No. P 31 46 706.7; Broadcast communications: Issue 1/1981, pages 1 to 6). So far it has been only developed a directional mixer that also on the free field transmission from Based on sound fields (13th Tonmeistertagung Munich 1984, pages 103 to 110). On the special importance of customization of the system "dummy headphones" is under different points of view (J. Blauert: Spatial Hearing, Postscript, new results and trends since 1972, S. Hirzel Verlag, 1985; Acustica: Vol. 48, Pages 272 to 274). In particular, to the 1 dB exact replica of the Polar pattern of the human ear pointed out in none of the prior art corresponding sound transmission method is observed. The measurement methods, the measurement setup and the results achieved with it in Connection with the determination of the Outer ear transfer function describes Blauert also at a glance (see above, Spatial hearing, postscript).

Modified, namely as a faster working and simplified procedure (for example lie after a few seconds of measurement already have graphics that draw conclusions about enable frequency-dependent distortions), F. König shows this with regard to the sound system variant "headphones" (patent application: P 39 03 246.9 and P 39 12 582.3).  

For recording and playback true to direction Public address systems were by Blauert / Boerger / Laws, Kürer / Plenge / Wilkens, Pleiderer etc. Efforts are underway to annoying side effect of "in-the-head localization" when playing sound recordings using Reduce headphone sound (patent registration: No. P 2 23 316.0, 26 28 053.0-31; Open documents: No. 19 27 401, 22 44 162, 25 45 446, 25 57 519; Funk Technik, issue 6 + 7/1984, Special printing).

However, it is necessary to use the defective ones Sound image reproduction properties of the head-related stereophony, in particular the recording-playback components individually to upgrade / adjust so that often at Listening to artificial head recordings, for example "In-head localization" to be registered is eliminated. In practice there is one special "front equalization" of artificial heads known, but only, statistically, a "stage effect" for the larger percentage generated by headphone sound. A generally dependent on the artificial head adapter Direction mapping "Front" or "Back" is here thus reduced to only "front". Furthermore is predominantly in theoretical papers a headphone customization discussed, but which in turn was conceived Analog filter circuit not for spatial Conditions but only for a sound direction of incidence on the artificial head, etc. applies.  

The production spread of headphones Membrane system transfer dimensions is the solution the location problems are irrelevant.

It would also make sense to focus on one the equalization variants mentioned with regard to the Transmission line artificial headphones too some by subjective evaluation standards should be exempt.

This task list to solve the problems in the unrealistic illustration of binaural Sound fields using headphones is used in a generic method by the characterizing features of claim 1 solved.

In order to properly implement the To ensure sound transmission process is for the transfer of the recording playback-side direction information a state of the art, compatible signal transmission method, preferably pulse code modulation (PCM), to use.

It is also advantageous that the receiving side a suitable dummy head and / or a Directional mixer that is the linear binaural Distortions and group delays Simulated artificial head, find use and whose analog stereo signals are preferably in digital PCM data are converted.  

It is also advantageous that a modified artificial head preferably over a switchable main and / or partial artificial head signal Directional analysis has either on the information content of both ears obtained, directional audio signal based or from a comparison from one to a spatial Zero degree deviation point to be obtained Sound source location, for example via light, infrared or electromagnetic Shaft-transmitter-receiver unit, determined or empirically by hand is entered.

It is also advantageous that the spatial directional analysis multidimensional, in a suitable resolution, preferably in a smaller one equal to 30 degree steps, and then a direction signal or direction Information in code form should correspond.

In addition, it is advantageous that parallel to simulate spatial positioning of mono signals using the directional mixer an associated direction signal or Direction information, preferably in one Resolution of less than or equal to 30 degree steps, is created.

Furthermore, it is advantageous that the Direction information preferably in one PCM code converted processed and saved becomes.  

It is also advantageous that the Directional information PCM code preferably the associated stereo PCM channels for one Audio signal transmission and / or storage is to be added on sound carriers.

Furthermore, it is advantageous that reproduced from the added Direction information PCM code using digital filter concept, a mode of operation of upper body, trunk, head and outer ear relief comprehensive so-called individual outer ear Directional equalization the digital-analog PCM audio signal to be converted is superimposed.

It is also advantageous that the Data for an individual outer ear Directional equalization from the difference of Outer ear transfer functions of the appropriate Dummy head and the individual person in one multidimensional resolution of preferably to take less than or equal to 30 degree steps are.

In addition, it is advantageous that for general signal transmission technical data of preferably at least 90 dB external voltage separation, one frequency response from 20 Hz to 20 kHz plus-minus 0.1 dB and a Distortion factor less than 0.01% are observed.  

It is also advantageous that for Transfer of direction information narrow band dips in suitable Listening areas, preferably at the top Hearing / transmission limit and / or in the C5 depression, using the "masking effect", too are realized.

Furthermore, it is advantageous that for Playback of the audio signals a suitable, on the artificial head and / or directional mixer coordinated, spatially true and / or sound reproduction equalized headphones use finds.

It is also advantageous that for a stereo sound signal recording the to be created, transferred and / or Direction information to be stored in discrete time intervals, preferably in steps of less than 100 milliseconds is collected / processed.

In addition, it is advantageous that for simultaneous multi-stereo channel pair transmission (stereo channel pairs: Ks _1,2 to Ks _{m, n} ; left channel: Ks ₁ to Ks _n-1 ; right channel: Ks ₂ to Ks _n ; n = m +1; n = m + 1; m = 1, 3, 5, 7, 9, ...) preferably each individual associated direction information (direction information: RI ₁ to RI _{n / 2} ) also transmitted or on Sound carrier is saved.

It is also advantageous to avoid of "transition claxes" when changing the playback Directional file (RI code change) a suitable sliding filter course change is created.  

It is also advantageous that at Sound carrier playback optionally between linear and directional information dependent and / or parallel to this free field, diffuse field and other equalized frequency responses switched can be.

The advantages achieved with the invention exist especially in that

• a) the existing, commercially available Sound recording and playback media such as for example compact disc player, only have to be upgraded,
• b) the one for a subsequent correction of the individual "directional Filters outer ear "(e.g. with artificial head Headphone system) as essential too transmitting direction information code of the entire usual audio listening frequency area just an imperceptible fraction, namely a bandwidth of less than 60 Hz takes up
• c) old recordings played unchanged / can be heard, i.e. it exists full compatibility with all stereo recording methods known today,
• d) an adaptation of the playback system to today common equalization (e.g. free field or Diffuse field) quickly and without further Additional devices can be implemented,
• e) which today mainly on a medium Directional characteristic of the artificial ear based recording technology, in interaction with headphone transducers, can be tailored to the individual,  
• f) individual timbres when listening to music deviations from the reference recording playback system (e.g. by the averaged anatomical Directional characteristic of an artificial head) be eliminated
• g) usual production variations in headphone Transducer of sound transducer, related to the "medium" directional characteristic e.g. one Artificial head, to be compensated (predominantly static distortion due to a constant direction of sound incidence towards the anatomical ear relief or buzzing effect of the headphone sound transducer outer ear PA characteristics),
• h) the subsequent necessary in some cases Equalization to an embossed and exclusive front location at Artificial heads are eliminated (as is well known the stage effect in the head-related Stereophonic), but in contrast all directional locations by means of the invention, especially in the horizontal and median level, possible are,
• i) the necessary, initially described by J. Blauert, 1 dB precise adjustment of the directional characteristic of the human ear to the realistic natural location of sound sources, related to the stereophonic Sound signal transmission, now when using the method disclosed in this invention, is also realized as well
• j) consequently the "in-the-head localization" known fault location of sound sources is turned off.

A practical embodiment of the invention this is based on the following figures and Describe explanations in more detail.

Show it

Fig. 1 shows two block diagrams showing the elements of a directional audio transmission coded, divided into recording (Fig. 1a) and reproducing area (Fig. 1b),

Fig. 2 is a diagram of the linear distortion of the "middle" outer ear when sound is "front" and

Fig. 3 is a diagram of the linear distortion of an individual outer ear when the sound hits "front".

In order to make the connections understandable, should be from one using a directional mixer generated binaural-natural stereophony went out. This could also be the case serve an artificial head, which is known to the same directional outer ear Has filtering properties like a Directional mixer. To simplify at this point, however, a sound signal recording and playback from a simulated acoustic ears to be located on both ears at the front Source for the demonstration station of the connections suffice that are characteristic of the process are.  

The recording part is first described in the block diagram according to FIG. 1a: With a microphone ( 1 ), a sample is taken from the low-frequency alternating signal provided by a sound source, converted into an alternating voltage and subsequently amplified ( 2 ) and fed to the directional mixer ( 11 ). In this directional mixer, the AC voltage is fed to the analog-digital converter ( 3 ), which enables pulse code modulation (PCM). The directional mixer ( 11 ) also has an input keyboard ( 5 ), so that the data entered into a corresponding digital filter curve ( 4 ) can be converted further in accordance with the "middle" directional characteristic of the outer ear, with the aid of a computer (6). The connection between a trigger generator ( 7 ), which, for example, should control the passage of newly received directional information every 100 milliseconds, and the alphanumeric keyboard ( 5 ) opens up a second possibility for entering directional information: it leaves replace the keyboard ( 5 ) with a transceiver (eg radio transmission), which is to be set up at a reference point. This provides "radar-like" from the movements of a non-fixed acoustic source (with the transmitter attached) an associated directional information (code).

Coming via the computer ( 6 ), triggered at the known time intervals, new equalizations are then generated in the filter unit ( 4 ). The direction-equalized signal (PCM data) thus leaving the filter unit is then linked in a suitable manner with the clocked direction information ( 9 ). The clocking is carried out by means of a digital memory switching element ( 8 ) which transmits the direction information provided by the computer ( 6 ) every 100 milliseconds.

In order to clearly illustrate the connection between the triggering and directional information, an implementation form that is unusual in practice, according to FIG. 1a, the connection of the filter unit ( 4 ), keyboard ( 5 ), data processing system ( 6 ), rectangle generator ( 7th ), Memory switching ( 8 ) and logic element ( 9 ) selected. The control of these elements ( 4 ) to ( 9 ) (except ( 6 )) and their logical-binary operations would take over a suitable computer-memory-bus concept in a finished device.

It should also be noted that the element shown as a digital filter ( 4 ) must be constructed much more complex than is symbolically shown in Fig. 1a, since our hearing decodes directions based on defined interaural group delays and equalizations. There are delays of approx. 0.5 milliseconds between the right and left ear signals, as well as damping and resonance due to the dimensions of the human anatomy.

The completed stereo digital signal can now be listened to by means of a digital-to-analog converter ( 10 ) or recorded in time using a tape recorder ( 12 ) or other data memories.

Today, a copy / reproduction (sound preserve) usually takes place on the medium compact disc ( 1 ). This process is associated with the method according to the invention in FIG. 1b. The transition between the recording ( Fig. 1a) and playback side ( Fig. 1b) can therefore be assumed to be fluid.

A compact disc player ( 7 ) adapted to the individual case works in the following way: The data read from the compact disc are split into directional information (RI) and stereo signal data. Via a connected keyboard ( 2 ) the RI can be let through or requested differently (keyboard) via data processing system (computer, ( 3 )) modified. The keyboard should preferably provide a ready-to-use free field and diffuse field equalization or filter-free transmission which, like the direction information code (RIC), triggers a defined effect in the digital filter ( 4 ). Further information on the practical implementation of a transmission (transmission bandwidth) of the necessary RIC data is given below in connection with the explanations relating to FIGS. 2 and 3.

The audio PCM signal, which is directionally equalized by means of the digital filter ( 4 ), is now conducted via a digital-to-analog converter ( 5 ) and a conventional preamplifier ( 6 ) to a power headphone amplifier ( 8 ).

A connected headphone ( 9 ) then sonicates an individual person ( 10 ). In comparison to the directional reference characteristic of a (statistically verified) "middle" ear, distortions of the headphone / sound transducer system and the anatomically present individual outer ear reefing occur. These frequency response differences (also the headphone transmission spread based on the reference headphones / transmission spread) apply depending on the direction of sound at the ear in e.g. 30 degree steps, from front to back, right, left, up and down, i.e. spatially assigned to record by measurement.

A realistic example for the case of a sound incidence direction "front" can be seen in the half-width pink noise bandpass analyzes, in Fig. 2 ("middle" ear) and Fig. 3 (individual ear). In particular, the striking relative half-ore analysis level differences above 2 kHz should be pointed out: In the "middle ear", a narrower and less pronounced depression around 8 kHz can be seen ( FIG. 2) than this in an individual person (cf. . Fig. 3) occurs. The same applies to the half-wide dip above 16 kHz. Furthermore, the frequency response increase by 5 kHz is a few decibels more pronounced in FIG. 3 than in FIG. 2. Below 2 kHz, no significant, therefore negligible frequency response differences can be found in FIGS. 2 and 3. They are based on the statistical behavior of the test signal "pink noise".

In contrast, all relevant level differences between the "middle" ( Fig. 2) and individual ear ( Fig. 3) are as level correction or compensation filter characteristics [with regard to the invention according to Fig. 1b, filter element ( 4 )] for the direction of sound incidence "in front "in the computer ( 3 ). They are called up and currently used when the direction information code "Front" is communicated by the compact disc player ( 1 ).

The calculation of the necessary transmission bandwidth of the directional information code can be carried out as follows: For example, a change in the directional information code is permitted every 100 milliseconds (t = 100 ms). The directional resolution is described with 30 degree steps in the horizontal and median plane (W = 2 _* {360/30} = 2 _* 12 = 24; W = number of varying values). In binary terms, there are 24 values <32 value = 2 ⁵ , thus 5 bits as a resolution in digital signal transmission. The system-theoretical digital transmission speed of information is V _s = 1 / t _s . 5 bits are transmitted in 100 ms, so there is time related to one bit equal to t _s = t / 5 bit = 100 ms / 5 bit = 20 ms per bit. From this follows V _s = 1 bit / 20 ms = 50 bits per second equal to V _s = 50 Bd (Boud). From this, the theoretical Nyquist bandwidth B _N = Vs / 2 = 50 Bd / 2 = 25 Hz can be calculated, from which the necessary system-theoretically defined transmission bandwidth of B _P = 1.4 _* B _N = 35 Hz results. In hi-fi technology, however, at least twice the sampling frequency (f _t <40 kHz; usually 44.1 kHz) of a low-frequency band range of 20 kHz is used, which results in the approximate directional information bandwidth of approx. 60 Hz results.

A withdrawal of 60 Hz at the top Transmission range limit frequency 20 kHz for the addition of the directional information code, would, for example, become one with the human senses do not hear noticeable from 19 980 Hz to 19 920 Hz reduced audio signal bandwidth (see 20 Hz up to 20 kHz with 20 Hz to 19 940 Hz).

Claims (16)

1. A method for fully directionally variable outer ear individual adaptation in head-related stereo sound transmission, characterized in that for the binaural individual adaptation of location-varying sound sources in suitable head-related sound transmission systems with the addition of a recording-side directional information for a reproduction-side outer ear transmission function coordination for realizing a natural multidimensional localization is made. 2. The method according to claim 1, d.g. that for the Transfer of the recording-reproduction side Direction information according to the state of the art Appropriate, compatible technology Signal transmission method, preferably Pulse code modulation (PCM), is used. 3. The method according to claims 1 and 2, d.g. that a suitable dummy head and / or a Directional mixer that is the linear binaural Distortions and group delays Simulated artificial head, find use and whose analog stereo signals are preferably in digital PCM data are converted.   4. The method according to claim 3, d.g. that a modified artificial head preferably over a switchable main and / or partial artificial head signal Directional analysis has either on the Information content of a two-eared, directional typical audio signal or from a comparison of one a spatial zero degree deviation point sound source location to be obtained, for example via light, infrared or electromagnetic wave transceiver unit realized, determined or empirically by Manual mode is entered. 5. The method according to claim 4, d.g. that the spatial directional analysis multidimensional, in a suitable resolution, preferably in a smaller one equal to 30 degree steps, and then a direction signal or direction Information in code form should correspond. 6. The method according to claim 3, d.g. that parallel to simulate spatial positioning of mono signals using the directional mixer an associated direction signal or Direction information, preferably with one Resolution of less than or equal to 30 degree steps, is created.   7. The method according to claims 5 and 6, d.g. that the direction information preferably in processed a PCM code processed and saved. 8. The method according to claim 7, d.g. that the Directional information PCM code preferably the associated stereo PCM channels for one Audio signal transmission and / or storage is to be added on sound carriers. 9. The method according to claims 1 to 8, d.g., that reproduced from the added Direction Information PCM Code a digital filter concept Mode of action of the upper body, trunk, head and so-called outer ear relief Individual outer ear directional equalization the digital-to-analog convertible PCM audio signal is to be superimposed. 10. The method according to claim 9, d.g. that the data for an individual outer ear Directional equalization from the difference of Outer ear transfer functions of the appropriate Dummy head and the individual person in one multidimensional resolution of preferably to take less than or equal to 30 degree steps are.   11. The method according to claims 1 to 10, d.g. that for general signal transmission technical data of preferably at least 90 dB external voltage separation, one frequency response from 20 Hz to 20 kHz plus-minus 0.1 dB and a Distortion factor less than 0.01% are observed. 12. The method according to claims 1 to 11, d.g. that to transmit the direction information narrow band dips in suitable Listening areas, preferably at the top Hearing / transmission limit and / or in the C5 depression, using the "masking effect", too are realized. 13. The method according to claims 1 to 12, d.g. that a suitable, for reproducing the audio signals on the artificial head and / or directional mixer coordinated, spatially true and / or sound reproduction equalized headphones use finds. 14. The method according to claims 1 to 13, d.g., that for a stereo sound signal recording the to be created, transferred and / or Direction information to be stored in discrete time intervals, preferably in steps of less than 100 milliseconds is collected / processed.   15. The method according to claims 1 to 14, d.g. that for simultaneous multi-stereo channel pair transmission preferably each and every associated one Transfer direction information with or is stored on sound carriers. 16. The method according to claims 1 to 15, d.g. that with sound carrier playback alternatively between linear and directional information dependent and / or parallel to this free field, diffuse field and other equalized frequency responses switched can be.