GB2367449A - Echo reduction in two-way communication - Google Patents

Abstract

A two-way communication device comprises a microphone, 4, and a loudspeaker, 8. The device determines the volume at which the device reproduces the signals it receives, e.g. by the setting of a volume control, 16, or via an energy level detector, 28. In order to reduce speech echo, means 10 and 14 (or 6 and 22) are provided for automatically adjusting the gain applied to the sound collected by the microphone, 4, in the opposite direction to changes the output speaker volume.

Description

2367449 Echo Reduction in Two-WaM Communication 5 This invention relates

to methods and apparatus for reducing echoes in two-way communications, particularly but not exclusively human voice communications over a radio link.

Many operating environments, especially for 10 portable radio communication devices, have high levels of ambient noise, e.g. building sites, airports etc. In such environments the received audio signal must be reproduced sufficiently loudly for the user to hear it.

However this also means that the propensity for the 15 reproduced audio to be picked up by the microphone is increased. If this happens the speaking party at the other end will hear his/her own voice echoed back, which is undesirable.

Several arrangements for reducing the amount of 20 echo have been proposed or used. For example digital signal processors have been used actively to cancel the echo element of a signal prior to its transmission.

Such a solution is however complex and expensive. More simply, the amount of echo can also be reduced by 25 physically separating the microphone and loudspeaker of the radio or the like from one another as much as possible. However the freedom to do this is often significantly restrained by the conflicting requirement, particularly for portable radios, to make the device as 30 compact as possible.

It is an object of the present invention to provide an improved solution and when viewed from a first aspect the invention provides a two-way communication device comprising a microphone and a loudspeaker, means for 35 determining the volume at which a signal received by the device is reproduced by the loudspeaker, and means for controlling the gain applied to sound collected by the microphone, said gain controlling means being arranged automatically to adjust said gain in the opposite direction to changes in the determined reproduction volume.

5 When viewed from a second aspect the invention provides a method of reducing echoes in a two-way communication system comprising automatically adjusting the gain applied to sound collected for transmission by a communication device, said adjustment being in the 10 opposite direction to changes in the volume at which that device reproduces received sound signals.

Thus it will be seen that in accordance with the present invention the gain applied -by the device to sound collected by the microphone is inversely dependent 15 upon a measure of the volume at which the loudspeaker reproduces received sound. This means that if the loudspeaker reproduces sound relatively loudly, the device will apply greater attenuation, or equivalently less amplification, to the sound collected by the 20 microphone, than if the loudspeaker reproduces sound relatively quietly. The result of this is the desired effect - namely that the degree of echo is reduced for a higher loudspeaker volume levels as compared to what it would have been if the microphone gain was fixed, 25 because less gain is applied to the sound picked up from the microphone and so it is relayed less loudly to the speaking party.

The actual value of the microphone gain could be positive - i.e. amplification, or negative - i.e.

30 attenuation, depending upon the particular circumstances. However only changes in this gain, rather than the actual value, are important in the present context.

In preferred simple embodiments, the means for 35 controlling the microphone gain is operatively coupled to a mechanism for controlling the volume at which the loudspeaker reproduces the sound signal it receives. In other words, the microphone gain is dependent upon the loudspeaker volume set by the user. The volume control will therefore control both the loudspeaker volume and microphone sensitivity, albeit in opposite directions.

5 Such an arrangement is convenient and simple, yet effective and thus when viewed from a further aspect the present invention provides a two-way communication device comprising a loudspeaker and a microphone, means for controlling the volume at which said loudspeaker 10 reproduces sound and means for applying gain to the sound collected from the microphone wherein said gain is determined according to the volume set on the volume control means.

For the reasons set out above it is preferred that is any change in the microphone gain is in the opposite direction to a change in the volume setting.

It will be apparent that by reducing the gain of the microphone audio path, the signal level received at the other end will be reduced. However it is observed 20 that when the operating environment is noisy, i.e.

exactly the circumstances in which a user is likely to increase the loudspeaker volume and thereby reduce the sensitivity of the microphone, users tend to raise their own voice level to compensate for the background noise.

25 The effect of this tends to counteract the effect of the reduction of microphone sensitivity. The overall effect is advantageous however since the effective signal-to noise ratio is improved.

Whilst the present invention is primarily 30 applicable to full duplex communication systems, i.e.

where both parties are able to transmit and receive simultaneously - which gives rise to the problem of echo - the embodiment described above may also be applicable to the systems such as half-duplex or push-to-talk 35 systems. The reason for this is that although such systems will not suffer from the sort of echo previously discussed, they may be able to benefit from the advantage given above that the signal-to-noise ratio in noisy environments will tend to be improved since the user will have the loudspeaker volume set high in these circumstances.

5 Changes in the loudspeaker volume and microphone sensitivity need not be in proportion. Indeed where, as is preferred, the gains of the loudspeaker paths and microphone are controlled by respective digital amplifiers/attenuators, the step sizes of the two may 10 not be the same. More generally the gain may be changed according to just one, or preferably more, thresholds in volume setting.

Preferably however the changes are in proportion or as near to being in proportion as is practicable. This is is advantageous since it means that the overall gain in the path from the received signal, via the loudspeaker and microphone, to the transmitted signal - and hence the level of echo will remain constant or at best substantially constant regardless of changes in volume.

20 By ensuring that this constant echo level is acceptably low, the amount of echo should always be within acceptable limits.

The means for controlling the loudspeaker volume may be embodied in any suitable way. For example it 25 could comprise a knob operating a potentiometer, or a pair of buttons respectively for signalling to a microprocessor to increase or decrease the volume. The volume control means could even be implemented in software - e.g. using voice activation.

30 As previously mentioned, coupling the microphone gain to the loudspeaker volume control setting is relatively simple and therefore inexpensive. In certain circumstances however there might be disadvantages to such an arrangement. For example if both parties to a 35 two-way call are in noisy environments and therefore both have their volume controls set high, the microphone gain at both ends, and hence the received audio level at the other end, will be reduced when each party needs it to be as high as possible.

Preferably therefore the microphone gain is additionally or alternatively dependent upon the actual 5 volume at which sound is reproduced by the loudspeaker.

This reproduction volume will of course depend upon the setting of the volume control, if any. It will, however, also depend on the level of the audio signal being received. Thus it will vary according e.g. to 10 whether the other party is speaking. There may also be other factors affecting the final reproduction volume. Monitoring the loudspeaker volume, however it is determined, gives the best performance in reducing the amount of echo since the reproduction volume is a direct 15 contributory factor to echo.

It will be seen that the advantage of this feature is that the amount of echo may be reduced as compared to known systems, but as long as only one party speaks at a time the microphone gain will not be undesirably 20 reduced. The reason for this is that the actual volume reproduced by the loudspeaker of the speaking party's device will not be high even if the volume control is set high since essentially it will only be reproducing background noise while the other party is not speaking.

25 The microphone gain will not therefore need to he reduced to avoid echo. The converse applies to the listening party and when the parties swap roles.

The present feature (of monitoring a measure of the actual volume of reproduced sound) could be seen as an 30 enhancement of the previous embodiment whereby the setting of the volume control and the level of the received audio signal could be measured and together used to determine the microphone gain.

Alternatively the sound itself could be measured 35 directly, e.g. using a small dedicated microphone. More preferably however the actual level of the audio signal relayed to the loudspeaker is monitored. This could be an analogue signal immediately prior to the loudspeaker, but is preferably a digital signal e.g. prior to conversion to an analogue signal for driving the loudspeaker. The latter possibilities will 5 automatically account for any other factors which might affect the reproduced volume and can be implemented relatively straightforwardly in hardware or software. For example a Digital Signal Processor could easily be used to monitor a digital domain loudspeaker signal.

10 The arrangements set out above are particularly convenient and thus when viewed from a further aspect the invention provides a two-way communication device comprising a loudspeaker and a microphone, means for monitoring a measure of the volume of sound reproduced is by the loudspeaker and means for applying gain to the sound collected from the microphone, wherein said gain is determined according to the reproduction volume determined by the volume determining means.

For the reasons set out earlier it is preferred 20 that any change in the microphone gain is in the opposite direction to a change in the volume setting.

Any change in reproduced volume level could he used to trigger a change in microphone gain. More preferably however one or more thresholds is applied. These could 25 he relative - e.g. percentage thresholds, but preferably a plurality of absolute thresholds is applied. The thresholds may be the same in each direction - or they may be direction dependent i.e. introducing hysteresis.

30 As previously discussed the changes in microphone gain need not be proportional to or even a function of changes in reproduced volume - they could e.g. just be predetermined for each threshold. of course where thresholds are employed, exact proportionality will not 35 be possible, but preferably the changes in reproduction volume and microphone gain are approximately in proportion, albeit in opposite directions so that, as explained before, the level of echo will be approximately constant.

The microphone gain could be arranged to change as soon as the reproduced volume changes or, as is 5 preferred, crosses a threshold. More preferably however a deliberate lag is introduced into the microphone gain changes. This prevents frequent changes in gain which could be disconcerting for a user. Most preferably the lag is applied to the reinstatement of gain following a

10 fall-off in the sound reproduction volume. This has been found to achieve good results since it allows a rapid gain reduction when the loudspeaker volume rises, thereby reducing the tendency to echo, whilst minimising rapid fluctuations in gain. Preferably a time constant 15 of approximately 0.1 to 0.5 seconds is used, most preferably approximately 0.25 seconds. This has been found to give adequate performance without interfering significantly with intelligibility during normal use.

The communication device according to the present 20 invention could comprise a fixed link, such as a wire or fibre optic link to another part of a communications network (which could simply consist of a similar device) Preferably however the device is cordless, most preferably portable. In the most preferred 25 embodiment the device comprises a mobile radio unit, e.g. a private mobile radio unit.

The method and apparatus of the invention may be implemented using pure hardware means such as discrete components or hard-wired logic gates. Preferably, the 30 invention is implemented at least partially using software, e.g. computer programs. It will thus be seen that when viewed from a further aspect, the present invention provides computer software specifically adapted to carry out the methods herein above described 35 when installed on data processing means.

Furthermore it will be appreciated that the means specified in the apparatus of the invention similarly preferably comprise computer software specifically adapted to carry out the methods herein above described when installed on data processing means, and a computer program element comprising computer software code 5 portions for performing the methods herein above described when the program element is run on a computer.

The data processing means and/or computer mentioned herein above is preferably comprised within the communications device which is most preferably a 10 portable device.

The invention also extends to a carrier comprising such software which when used to operate a two-way communication device comprising a digital computer, causes, in conjunction with said computer, said device 15 to carry out the steps of the method of the present invention. Such a carrier could be a physical storage medium such as a ROM chip, CD ROM or disk, or could be a signal such as an electronic signal over wires, an optical signal or a radio signal such as to a satellite 20 or the like. Preferably the carrier comprises a microprocessor chip.

It will further be appreciated that not all steps of the invention need be carried out by computer software and thus from a further broad aspect the 25 present invention provides computer software for carrying out at least one of the steps of the methods set out herein above when used to operate means for carrying out the other steps. Similarly, not all of the means specified in the apparatus of the invention need 30 comprise computer software and thus in the general preferred case, it is at least one of such means which comprises computer software.

Of course it should be understood that the terms "loudspeaker" and "microphone,, are used generically to 35 cover means for reproducing and collecting sound respectively, and no particular construction or method of operation is to be inferred from these words.

In the preferred embodiment the loudspeaker comprises a loudspeaker of known type and similarly the microphone comprises a microphone of known type.

Certain preferred embodiments of the invention will 5 now be described, by way of example only, with reference to the accompanying drawings in which:

Figure 1 is a schematic block diagram of a two-way radio in accordance with a first embodiment; and Figure 2 is a schematic block diagram showing the 10 operation of a two-way radio in accordance with a second embodiment.

A schematic block diagram of a portable private two-way mobile radio unit, in accordance with a first preferred embodiment of the invention, is shown in 15 Figure 1. The radio unit is full duplex - i.e. both parties can transmit and receive simultaneously.

The basic operation of such radio units is well known to those skilled in the art and will not, therefore, be explained in detail here. Briefly, 20 however, the receiver 2 of the unit receives a radio signal R which it processes by performing any necessary demodulating, decoding, etc. The resultant audio signal extracted is passed to a loud speaker 4 via a variable attenuator 6. The attenuator 6 thus controls the volume 25 at which the loudspeaker 4 reproduces the received sound signal.

on the transmission side, a microphone 8 collects sound and converts it to an electrical signal. This electrical signal is passed via a second variable 30 attenuator 10 to a transmitter unit 12. The transmitter unit performs the reverse of the operations performed by the receiver unit 2 and thus, codes the signal, modulates it, etc. The transmitter then transmits a radio signal T to another part of the network - e.g., a 35 base station or a similar mobile radio unit.

Both the loudspeaker attenuator 6 and the microphone attenuator 10 are controlled by a micro- controller 14 which comprises a digital micro-processor.

In common with known radio units, the micro-controller takes an input signal from a volume control knob 16 which it uses to determine the degree of attenuation to 5 apply to the loudspeaker signal by the attenuator 6.

The volume knob, shown schematically at 16, is of the rotary potentiometer type in which the form of the signal supplied to the micro-controller is a variable voltage according to the resistance of the 10 potentiometer. Equally however, the volume control 16 could be a pair of buttons respectively for increasing and decreasing the volume.

Echo will be produced if sound reproduced by the loudspeaker 4 is picked up by the microphone 8. In 15 known arrangements the attenuator in the microphone path would have been fixed at the time of manufacture so as to give acceptable echo at ordinary listening volumes.

Inevitably this would result in increased echo if the volume setting was increased.

20 In contrast with known radio units, however, the micro-controller 14 of the present embodiment also varies the gain in the audio path from the microphone 8 to the transmitter 12 by adjusting the setting of the attenuator 10. Thus in this embodiment, the tendency 25 for the radio unit to produce echo is reduced since the micro-controller 14 is arranged to make equal, but opposite, adjustments to the two attenuators 6, 10 when the volume knob 16 is adjusted.

Thus, if the setting of the volume knob 16 is 30 increased, the micro-controller 14 will accordingly increase the gain in the audio path between the receiver 2 and the loudspeaker 4 by reducing the amount of attenuation applied by the attenuator 6. At the same time, however, it will increase the amount of 35 attenuation applied by the attenuator 10 to sound collected by the microphone 8 and which is transmitted by the transmitter 12. Thus, the potential tendency for echo to be increased when the setting of the volume knob 16 is increased by virtue of the greater volume at the loudspeaker 4 and therefore, at the loudspeaker 8, is avoided by reducing the sensitivity of the microphone 8 5 by a corresponding amount, Similarly, if the setting of the volume knob 16 is reduced, not only will the micro-controller 14 increase the attenuation applied by the attenuator 6 and therefore reduce the volume of the sound reproduced by 10 the loudspeaker 4, it will also decrease the attenuation applied by the attenuator 10 by corresponding amounts, thereby enhancing the sensitivity of the microphone 8 without creating any greater tendency for producing echo. Thus, by carefully setting up the mobile radio 15 unit and, in particular, the absolute values of the attenuation applied by the two attenuators 6, 10 to give an acceptably low level of echo, the degree of echo can be maintained within acceptable limits, regardless of the setting of the volume knob.

20 Turning now to Figure 2, there will be seen a schematic block diagram showing the operation of a second embodiment of the invention. As before, this is also a two-way, full duplex, private mobile radio unit.

Considering firstly the transmission side of the 25 unit, sound is picked up by the microphone 8 and converted to an analogue electrical signal. This analogue signal is converted to a digital one by an, analogue-to-digital converter 18. The signal is passed to a transmission unit 20 via a variable attenuator 6 30 which controls the gain applied to the signal. The attenuator 6 is controlled by a gain control unit 22 as will be explained in greater detail hereafter.

On the reception side, the received and processed digital audio signal 24 is passed to a digital-to 35 analogue converter (DAC) 26 and the resulting analogue signal is used to drive the loudspeaker 4. However, immediately prior to the DAC 26 is an energy level detector 28. This detector 28 passes a signal corresponding to the detected energy level to a filter module 32. The filter module passes an energy value to a threshold comparator 30. The threshold comparator 30 5 passes a signal to the gain control unit 22.

The operation of this system is as follows. The energy level detector 28 measures the signal level of the driving signal 24 for the loudspeaker 4 immediately prior to its conversion to analogue by the DAC 26. This 10 signal level therefore gives a.good measure of the volume at which the loudspeaker 4 will reproduce the sound.

The energy level measurement is taken by the detector 28 over a short time period of 7.5 ms and is 15 therefore effectively instantaneous. This instantaneous energy (IE) value is passed to a filter unit 32 where it is compared with a previously stored value for the mean Long Term Energy (LTE). If the IE value is greater than the LTE value, the latter is used to overwrite the LTE 20 value in memory. If the IE value is lower than the stored value, no action is taken at this point.

The IE value just measured is then passed onto the threshold comparator 30 via a low-pass filter within the filter module 32 in the form of an exponential damper 25 having a time constant of 250ms. The filter takes as its input the IE value and its output is used to modify the stored LTE value. The LTE value is then passed onto the threshold comparator.

Clearly if the IE value was higher than the 30 original LTE value, the LTE will have been deliberately set to be the same as the IE and the filter will thus have no effect meaning that the new IE value will be passed straight onto the threshold comparator 30.

If however the original LTE was higher, the filter 35 will cause a damped response in the LTE value towards the lower IE value. In a simple example if the IE remains constant the LTE will decrease exponentially so that the difference is halved every 250 milliseconds (the time constant of the filter).

Effectively therefore the filter unit 32 gives a step response in output when the energy of the signal 24 5 rises, but when the energy falls, the filter's response is exponentially damped with a time constant of approximately 250 milliseconds.

The comparator 30 compares the energy value to four predefined thresholds. It then passes a signal the gain 10 control unit 22 depending on which of the thresholds have been exceeded. Each threshold corresponds to -3 dB of attenuation. Thus if the other party to a conversation is not speaking the energy may exceed none of the thresholds, in which case no attenuation will be 15 proscribed by the gain controller 32, or if the volume knob is set high the background noise may cause the first threshold to be exceeded, in which case an attenuation of -3 dB will be proscribed. on the other hand, if the other party is speaking and the volume knob 20 is, say, set high, it is likely that all four thresholds will be exceeded and therefore a maximum -12 dB will be proscribed.

The determined level of gain is signalled to the attenuator 6 by the gain controller 22.

25 Thus it will be seen from the foregoing description that the microphone gain set by the attenuator 6 is varied oppositely to changes in the volume at which the loudspeaker 4 reproduces sound (as measured by its digital driving signal). The increase in tendency for 30 echo to arise as sound is reproduced more loudly by the loudspeaker 4 is therefore countered by a reduction in the sensitivity of the microphone 8.

Moreover by virtue of the asymmetric response of the filter unit 32 (resulting from the IE value being 35 used to replace the LTE value in the event that the former is greater), the unit will respond rapidly to rises in the level of the audio signal 24 and thus to the volume reproduced by the loudspeaker 4, by reducing the gain of the microphone (i.e. increasing the attenuation applied by attenuator 6.) In the opposite case however, where the energy level falls, the energy 5 value which is compared to the thresholds responds more slowly and thus the unit will increase the gain (reduce the attenuation) only slowly if the reproduced volume falls. This prevents rapid fluctuations in the microphone gain whilst effectively preventing echoes as 10 quickly as possible.

Although the various modules such as the energy level detector 28, threshold comparator 30 and filter module 32 etc. have been shown as separate blocks, this is merely for the purposes of clarity of explanation.

15 In fact all of these functions are in fact implemented by suitable software running on a microprocessor in the unit. Some or all of the functions could equally be implemented in hardware.

Claims (20)

Claims:

1. A two-way communication device comprising a microphone and a loudspeaker, means for determining the 5 volume at which a signal received by the device is reproduced by the loudspeaker, and means for controlling the gain applied to sound collected by the microphone, said gain controlling means being arranged automatically to adjust said gain in the opposite direction to changes in the determined reproduction volume.

2. A device as claimed in claim I wherein the means for controlling the microphone gain is operatively coupled to a mechanism for controlling the volume at is which the loudspeaker reproduces the sound signal it receives.

3. A two-way communication device comprising a loudspeaker and a microphone, means for controlling the volume at which said loudspeaker reproduces sound and means for applying gain to the sound collected from the microphone wherein said gain is determined according to the volume set on the volume control means.

4. A device as claimed in claim 3 wherein said means for controlling loudspeaker volume and means for applying gain are arranged such that any change in the microphone gain is in the opposite direction to a change in the volume setting.

5. A device as claimed in any preceding claim arranged such that changes in loudspeaker volume and microphone gain respectively are in proportion or as near to being in proportion as is practicable.

6. A device as claimed in any preceding claim wherein the microphone gain is dependent upon the actual volume - 16 at which sound is reproduced by the loudspeaker.

7. A device as claimed in claim 6 comprising means for measuring the actual level of the audio signal relayed 5 to the loudspeaker.

8. A device as claimed in claim 7 wherein said means for measuring is arranged to measure a digital signal.

10

9. A two-way communication device comprising a loudspeaker and a microphone, means for monitoring a measure of the volume of sound reproduced by the loudspeaker and means for applying gain to the sound collected from the microphone, wherein said gain is 15 determined according to the reproduction volume determined by the volume determining means.

10. A device as claimed in claim 9 wherein said means for applying gain is arranged to change the gain applied 20 in the opposite direction to changes in the determined volume.

11. A device as claimed in any preceding claim wherein said means for applying gain is arranged to change the 25 gain applied only if the volume crosses one of a predetermined plurality of thresholds.

12. A device as claimed in any preceding claim wherein said means for applying gain is arranged to change the 30 gain only after a time delay.

13. A device as claimed in any preceding claim comprising a mobile radio unit.

35

14. A method of reducing echoes in a two-way communication system comprising automatically adjusting the gain applied to sound collected for transmission by 17 - a communication device, said adjustment being in the opposite direction to changes in the volume at which that device reproduces received sound signals.

5

15. Computer software specifically adapted to carry out the method of claim 14 when installed on data processing means.

16. A two-way communication device comprising software 10 as claimed in claim 15.

17. A carrier comprising software which when used to operate a two-way communication device comprising a digital computer causes said device to carry out the is steps of the method of claim 14.

18. A two-way communication device substantially as hereinbefore described with reference to the accompanying drawings.

19. A method of reducing echoes substantially as hereinbefore described with reference to the accompanying drawings.

25

20. Computer software substantially as hereinbefore described with reference to the accompanying drawings.