GB2370176A - A simple microphone unit for the vertical localisation and enhancement of live sounds - Google Patents

Abstract

A vertical localisation microphone unit 10 that is capable of capturing the height of a sound 15 during a recording, so that on playback, the vertical dimensional quality of this sound 15 can be reproduced. The unit 10 contains two microphones 11, 12 which are positioned apart vertically and horizontally at distances A and B respectively. Their outputs are mixed 13 and passed for delivery 14 to possibly an amplifier or recording device. The position of the microphones 11 and 12, the angle of the incident sound source 15, and the summation of their output signals is a non-ideal copy of the delay-and-add mechanism believed to be employed by the human pinna for vertical sound localisation. This unit 10 requires the use of only a single channel to deliver the reproduction of height in a sound, so no modification to existing equipment is required.

Description

A SIMPLE MICROPHONE UNIT FOR THE VERTICAL LOCALISATION AND ENHANCEMENT OF LIVE SOUNDS This invention relates to a microphone unit that aims to enhance the capabilities and performance of existing live recording equipment. It is able to capture the sense of height from a live sound by generating and incorporating artificial vertical localisation cues into the signal output, which can then be recorded as usual on a mono channel.

When the signal is played back through loudspeakers or preferably headphones to a listener, the cues are naturally interpreted and understood, allowing vertical discrimination of sounds and making them much more realistic.

The invention needs only one channel, using two microphones in combination to simulate the vertical localisation cues normally imparted to live sounds by the human pinna. The invention requires no special recording and playback equipment, and when used with normal stereo equipment, sounds will have both vertical and horizontal direction.

Obvious applications for this technology are the music and film industry where a more life-like sound recording and presentation will enhance and improve the quality of the material being presented.

This unit could also have an application in hearing aids. Users currently have difficulty filtering out background noise, mainly because they are unable to determine and focus on the source of sound they wish to listen to. This microphone device would provide vertical localisation cues to allow the listener to locate and focus on sound sources in the vertical plane. However, for improving horizontal localisation, a unit for each ear may be necessary (i. e. binaural).

Another application could involve the use of the device to hear sounds in water, possibly for divers to locate sound sources. Sound travels too fast in water for the ears to generate the right vertical localisation cues, so this invention could be scaled up to artificially generate the right cues. Divers listening to the output of the unit could then identify the vertical position of sounds, but to locate sounds in the horizontal plane, stereo recording technology may also be required.

Currently, the most successful technological form for improving the recording and playback of sound is known as stereophonic or"stereo"technology. Improvement is achieved by the use of two microphones, spaced a certain distance apart along the horizontal. Two record and playback channels are required, but the result is the delivery of a much wider sound than that given from a single channel.

A more advanced form of stereophonic recording technology was introduced called 'quadraphonic'technology, which employed the use of four channels instead of two.

Four microphones were required, two for the left and right in front and another pair for the left and right behind. This technology was never fully commercially successful, mainly because the sound improvement was not good enough for the need of four microphones and four speakers.

A microphone technology which aims to offer full three dimensional sound recording is one known as binaural recording. This techniques uses a copy of a typical human head and pinnae, with microphones positioned in the ear canals. Any sound reaching the microphones will contain basic localisation cues (due to the influence of the head and pinnae), which can be interpreted by a listener during reproduction. The recorded sounds can be delivered through headphones for best effect, or through loudspeakers with limited effect.

Although binaural recording is effective in gaining a more realistic recording, it suffers in two main areas: it can be impractical to use (as can quadraphonics), and is aesthetically unpleasant. This is believed to have prevented the commercial success of this recording technology.

Background theory, and two specific embodiments of the invention will now be presented by way of example with reference to the accompanying drawings in which: Figure 1 shows the basic delay-and-add model around which the invention is based.

Figure 2 illustrates how a human pinna creates a direct and delay path in a similar way to the delay-and-add model.

Figure 3 is a graphical representation of the spectral boosting and attenuating of frequencies influenced by a human pinna at different elevation angles.

Figure 4 shows the two microphone set-up of the preferred embodiment incorporating an acoustic delay for the x-and z-axis delay.

Figure 5 shows the two microphone set-up of the second embodiment incorporating an acoustic delay for the z-axis delay and an electronic, electromechanical or electroacoustic delay mechanism for the x-axis delay.

Figure 6 is a graph showing the closest match of the angle-delay responses of the two embodiments to an experimental representation of how a human listener may position a sound with varying delays using the delay-and-add model (reference only for-200 to +40 perceived elevation).

A human's auditory system can provide information on the location of a sound source to the brain in three dimensional space using three different methods. Two methods are used to locate the sound in the horizontal plane, but the method for positioning in the vertical plane can be more complex.

Localisation in the horizontal plane is achieved by the brain comparing the amplitude and/or phase differences between sound waves reaching the two ears. Stereo microphone units use these two principals to capture the breadth of a sound.

The same methods for horizontal localisation may not be used for vertical localisation since there is no vertical displacement between the ears. Instead, it is believed that the shape of the ear lobe or pinna is mainly used to subtly change the spectral dimensions of the incoming sound depending on its vertical elevation.

For sound sources outside the angles-200 and +90 elevation and behind the listener, the mechanisms for generating the spectral changes can be very complex, but inside these angles, the mechanisms are much simpler. It is believed that the ear uses a delay-and-add model as shown in Figure 1 to alter the sound's spectral shape. This shape is then used by the brain to estimate the elevation of a sound source.

For the localisation of a sound source by a human listener, the external pinna creates both a direct and lengthened (or delayed) path for the sound reaching the ear drum as shown in Figure 2, with the lengthened path being generated through reflection off the concha. However, due to the shape of the concha, the length of this path is proportional to the incident angle of the sound, i. e. this length is longer for negative angles and shorter for positive angles. The sounds travelling the two paths are then summed at the opening of the ear canal, altering the sound's spectral shape before it is passed to the inner ear. These shapes are illustrated in Figure 3 for three different elevation angles :-20 , 0 , and +40 . It is believed that this spectral shape is decoded by the brain using a delay subtraction method to identify the delay length and therefore the vertical location of the sound source.

So for a listener localising a sound source in the vertical plane, the sounds spectral shape is altered through incident angle dependant acoustic filtering, which is then decoded by the brain to identify the sounds position.

It should be noted that for good vertical localisation, the sound must have a reasonably high bandwidth, preferably between 5k and 20kHz, but this is normal for every-day sounds and most types of music.

The preferred embodiment of my invention shown in Figure 4, consists of a microphone unit 10 having two microphones 11, 12 facing toward a sound source 15, which are positioned in space in such a way that the time taken for a sound to travel

from microphone 11 to microphone 12 along the horizontal B is in the region of 215Jls, and along the vertical A in the region of 78us. In air, these distances could be B=74mm and A=27mm, but will change depending on the medium through which the sound is travelling. The outputs of the microphones are then mixed 13 (optimal 1 : 1), and the output 14 is delivered for recording purposes.

The acoustic delay lengths A and B are used to geometrically alter the time delay between signals reaching microphones 11 and 12 with respect to the incident angle.

For example, for a sound source at a positive elevation angle, the extra distance the

sound would travel to reach microphone 11 after 12 would be less than if the sound source was at a negative angle (but -20 ). This technique is similar to the way the concha is used to create different delay lengths prior to summation at the ear canal as shown in Figure 2.

It should be noted that the acoustic delays or distances A and B are important as any deviations could affect the spectral pattern of the output signal and reduce the effect of the microphone system.

The embodiment described here does not closely match the delays influenced by the pinna with respect to elevation for large angles as well as the second embodiment of Figure 5, but was chosen as the preferred due to it's simplicity for practical realisation and commercial exploitation. The angle to delay response of this model can be seen in Figure 6, which is compared against an experimental representation of how a human listener may position a sound with varying delays using the delay-and-add model.

It should also be noted that sounds picked up by the microphones at angles outside - 200 and +90 elevation will not be represented accurately, so to reduce their effect, it is recommended to use directional microphones with their sensitivities maximised within these angles.

The second embodiment shown in Figure 5 contains a microphone unit 20, but the horizontal acoustic delay B of Figure 4 is replaced by an electronic, electromechanical or electroacoustic delaying device 23, incorporating the same time delay B in the region of 215 us. This allows microphone 21 to be mounted directly above microphone 22, at a distance A in the region of 78 us, with the output of microphone 21 feeding into the delaying mechanism 23. The output of this mechanism is then mixed 24 with the output of microphone 22, and the output 25 fed for reproduction purposes.

It can be seen from the graph of Figure 6 that the angle to delay response of this embodiment matches the human interpretation response more closely than the preferred embodiment. However, this embodiment requires the addition of an electronic, electromechanical or electroacoustic delaying mechanism to function properly.

Claims (20)

1. CLAIMS 1. A method of recording using two spaced microphones wherein the two microphones are positioned in a common physical plane which passes through the sound source.
2. 2. A method according to claim 1 wherein the outputs of the two microphones are added in a mixer and the connections of these microphones to the mixer are arranged so that there is a time delay between the arrival of the sound signals at the mixer.
3. 3. A method according to claims 1 and 2 wherein the mechanism for introducing a time delay is a physical distance along which the acoustic sound travels.
4. 4. A method according to claims 1, 2 and 3 wherein the mechanism for introducing a time delay is an electronic signal delay.
5. 5. A method according to claims 1,2, 3 and 4 wherein the time between signals arriving at the mixer will vary in proportion to the physical position of the sound source in the physical plane.
6. 6. A method according to claims 1,2, 3,4 and 5 wherein the time delay between the signals reaching the mixer for a particular sound source position is substantially close to the time delay required to position the same sound using the'delay-and-add'model (which has been proposed for vertical sound localisation).
7. 7. A method according to claims 1,2, 3,4, 5 and 6 wherein the perceived effective vertical localisation angle with respect to 0 degrees at the horizontal can be substantially between-40 and +90 degrees.
8. 8. A method according to claims 1,2, 3,4, 5 and 6 wherein the mixing conditions can be altered to reduce the localisation effect by attenuating one of said mixer inputs.
9. 9. A method according to claims 1,2, 3,4, 5 and 6 wherein the bandwidth of the sound source to be perceived is between 5 kHz and 20 kHz for effective localisation.
10. 10. A method substantially as described with reference to any one of accompanying drawings.
11. 11. Apparatus for recording including two microphones which are positioned in a common physical plane which is adapted to pass through the source of sound that is to be recorded.
12. 12. Apparatus according to claim 11 wherein the outputs of the two microphones are added in a mixer and the connections of these microphones to the mixer are arranged such that there is a time delay between the arrival of the sound signals at the mixer.
13. 13. Apparatus according to claims 11 and 12 wherein the mechanism for introducing a time delay is a physical distance along which the acoustic sound will travel.
14. 14. Apparatus according to claims 11,12 and 13 wherein the mechanism for introducing a time delay is an electronic signal delay.
15. 15. Apparatus according to claims 11,12, 13 and 14 wherein the time between signals arriving at the mixer will vary in proportion to the physical position of the sound source in the physical plane.
16. 16. Apparatus according to claims 11,12, 13,14 and 15 wherein the time delay between the signals reaching the mixer for a particular sound source position is substantially close to the time delay required to position the same sound using the'delay-and-add' model (which has been proposed for vertical sound localisation).
17. 17. Apparatus according to claims 11,12, 13,14, 15 and 16 wherein the perceived effective vertical localisation angle with respect to 0 degrees at the horizontal can be substantially between-40 and +90 degrees.
18. 18. Apparatus according to claims 11,12, 13,14, 15 and 16 wherein the mixing conditions can be altered to reduce the localisation effect by attenuating one of said mixer inputs.
19. 19. Apparatus according to claims 11, 12, 13, 14, 15 and 16 wherein the bandwidth of the sound source to be perceived is between 5 kHz and 20 kHz for effective localisation.
20. 20. Apparatus substantially as described with reference to any one of accompanying drawings.