WO1999011047A1

WO1999011047A1 - Method and apparatus for listener sidetone control

Info

Publication number: WO1999011047A1
Application number: PCT/CA1998/000674
Authority: WO
Inventors: Christopher Michael Forrester; John Stephen Graham; Paul Provencal
Original assignee: Northern Telecom Limited
Priority date: 1997-08-21
Filing date: 1998-07-22
Publication date: 1999-03-04

Abstract

A method of adjusting listener sidetone in a handset is provided. Measured ambient noise level of the handset microphone signal is compared against a predetermined threshold. The sidetone level is adjusted by a predetermined amount in response to whether the measured ambient noise level received at the handset microphone is higher or lower than the predetermined threshold.

Description

METHOD AND APPARATUS FOR LISTENER SIDETONE CONTROL

Field of the Invention

This invention relates to sidetone control in a telephone apparatus and

more particularly, to a method of reducing listener sidetone in a noisy

environment.

Background of the Invention

Sidetone is defined as a small amount of transmitter signal which is fed

back into the talker's receiver. A small amount of sidetone is desirable in the

telephone so that the user can hear his or her own voice and therefore

determine how loudly to speak.

The sidetone must be at a proper level. Too much sidetone will cause a

person to speak too softly for good reception by the far-end party.

Conversely, too little sidetone will cause a person to speak so loudly that it

may sound like a yell at the other end.

Unfortunately, the introduction of sidetone has been found to be

problematic in an environment where ambient noise is high. Local ambient

noise, such as generated in noisy road intersections, airports, factory floors

etc., is coupled into the receiver channel by way of the sidetone path. The

louder the ambient noise, the more difficult it is for the telephone user to hear the other party's voice. Users complain that they "can't hear" the far end

talker under these conditions. To minimize this, it is desirable to reduce the

noise fed back through the sidetone path by lowering sidetone loudness.

This, however, should only be done in high noise environments, since a

nominal sidetone level is required for user comfort in quieter environments.

The amount of ambient noise fed back to a user via the sidetone path in

a telephone set is referred to as Listener Sidetone Rating or LSTR.

Specifications for different telephone systems typically require that LSTR be

greater than 15 dB. Unfortunately, most handset terminals do not meet this

requirement. In high noise environments, sidetone levels can be reduced

since the user will or should speak louder.

The need therefore exists for a method and apparatus for controlling

and reducing listener sidetone in a telephone apparatus.

Summary of the Invention

It is therefore an object of the present invention to provide a method of

dynamically reducing listener sidetone according to the noise or ambient

noise level detected at the handset microphone.

Another object of the present invention is to provide a method and

apparatus for controlling listener sidetone wherein if the noise level is sufficiently high, a loss is inserted in the sidetone path to improve LSTR

performance.

According to an aspect of the present invention, there is

provided a method of adjusting listener sidetone comprising the steps of:

measuring the ambient noise level of a received input signal;

comparing the measured ambient noise level against a

predetermined threshold; and

adjusting the sidetone level by a predetermined amount in

response to whether the measured ambient noise level is higher or

lower than said predetermined threshold.

According to another aspect of the present invention, there is

provided a method of adjusting listener sidetone in a handset, comprising the steps of:

measuring the ambient noise level of the handset microphone

signal;

comparing the measured ambient noise level against a

predetermined threshold; and

adjusting the sidetone level by a predetermined amount in

response to whether the measured ambient noise level received at the handset

microphone is higher or lower than said predetermined threshold. Brief Description of the Drawings

Fig. 1 is a schematic diagram illustrating generally how sidetone is

generated in a circuit of the prior art;

Fig. 2 is a schematic diagram illustrating the basic elements of a

controllable sidetone /LSTR path according to an embodiment of the present

invention; and

Fig. 3 is a flow diagram illustrating the listener sidetone controlled

algorithm according to a preferred embodiment of the present invention.

Description of the Preferred Embodiment

Referring now to Figure 1, we have shown a basic schematic diagram

illustrating how sidetone is provided in a typical telephone handset. The

telephone handset is depicted by microphone 11 and receiver 12. A sidetone

path 13 is established between the microphone and receiver ends of the

handset. Sidetone loss is inserted in the sidetone path by means of an

attenuator 14. The attenuator 14 is typically a resistive voltage divider

network but it could also be an operational amplifier. A predetermined

adjustment is made to attenuator 14 to fix the level of the sidetone received at

the receiver 12. For users, the sidetone is fixed at one value for all conditions,

thus LSTR is also fixed for all room noise conditions.

Shown in Figure 1 is a network echo path 15. This echo occurs at

network interfaces such as the four-two-wire hybrid in an analog telephone set or the point of interconnect of a wireless phone network into the telephone

network for example. When the delay in this echo path 15 is very short, i.e.

less than 20 milliseconds, it will sound like sidetone to listeners. When the

delay is very long, i.e. greater than 20ms, this echo disrupts normal

communication. This path must therefore be eliminated. Elimination of the

network echo signal is accomplished using echo cancellation techniques either

in the telephone terminal or within the telephone network. These techniques

are well known to those skilled in the art. Since it is assumed that this echo

has been effectively eliminated, the network echo 15 is not shown in remaining figures.

Referring now to Figure 2, we have shown the basic elements forming

part of the present invention and, in particular, the method and apparatus for

controlling sidetone/LSTR paths. As in the embodiment of Figure 1, a fixed

sidetone level loss is introduced in the sidetone path 23 by means of a fixed

attenuator 24. However, with the apparatus of the present invention, the

overall sidetone level loss introduced in the sidetone path 23 is dynamically

adjusted according to the ambient noise level measured at the microphone 21.

The overall sidetone level loss introduced in the sidetone path 23 is achieved

using a fixed attenuator 24, providing a fixed sidetone level loss, and a

dynamically variable attenuator 25, providing a variable sidetone loss.

Typically, the variable attenuator 25 is implemented using a programmable loss amplifier, which is controllable via a micro-controller serial interface or

an analog DC voltage.

In order to determine the amount of additional sidetone loss to be

introduced in the sidetone path, the level of ambient noise introduced at the

microphone 21 needs to be estimated. The output signal 22 of the microphone

21 is converted from to a digital signal format via the analog-to-digital

converter 30. The ambient noise picked-up at microphone 21 is estimated by

means of a microphone noise estimator 28. This is done by appropriately

discriminating noise from speech so that a true ambient noise level can be

estimated. Noise estimation techniques are well known to those skilled in the

art. For example, a classic noise estimator consists of a full-wave rectifier

followed by slow attack and fast decay processing of the signal.

Although the embodiment shown in Fig. 2 makes use of a

microcontroller based implementation, the invention described herein can also

be implemented using an analog implementation to carry out the estimation

of the ambient noise and the LST algorithm.

In addition, although the illustrated embodiment is implemented in an

analog telephone set, the circuitry can be adapted to operate with a digital

telephone set. In this case A/D and D/A converters would be used between

the transducers and the network. As indicated above, the microphone noise estimator 28 is used to

estimate the ambient noise levels picked-up at the microphone. The LST

controller 29 makes use of the estimated noise level to determine how much

loss is to be inserted in the sidetone path. In the sidetone path, the variable

attenuator 25 is then adjusted according to whether additional sidetone level

loss is increased or decreased according to the computations made by the LST

controller 29. The insertion or removal of additional sidetone loss is done at

rates that cause the least noticeable amount of "noise pumping" or other audio

artefacts, but keeps LSTR performance optimized for changing environmental

noise. The rate of insertion or removal of the additional sidetone loss is also

referred to as the attack and decay rates of sidetone loss.

As indicated earlier, the sidetone path 23 also includes a fixed sidetone

level provided by attenuator 24 which is present to ensure that nominal

sidetone levels are available for normal call conditions and provides a fixed

loss in the sidetone path from which to start the listener sidetone control.

This fixed loss is non-dynamically adjusted to optimise the best case sidetone.

In operation, if the handset is used in a noisy environment, the

microphone noise estimator is used to determine the ambient noise level

present at the microphone 21. If the noise level is sufficiently high, an

additional amount of sidetone loss is inserted in the sidetone path to improve

LSTR performance. If noise levels are sufficiently low, the additional amount of sidetone loss is removed in the sidetone path to optimize sidetone

performance.

Referring now to Figure 3, we have shown a flow diagram illustrating

how the LST controller monitors microphone noise levels to adjust the

sidetone loss level of the sidetone path. Upon initialization 30, the attack and

decay timer values are set to predetermined values, block 31. As a first step to

the LST control process block 32, the microphone noise level is retrieved from

the microphone noise estimator 28 of Figure 2. The computed microphone

noise level is used to determine the need for adjusting sidetone path loss. At

box 34, if the microphone noise is less than a predetermine threshold, the

target sidetone loss is computed to remain at a standard base level, block 35.

That is, no additional sidetone loss is the target for the sidetone path.

However, if the microphone noise is greater than the threshold, a target

sidetone loss is computed, block 36, according to a predetermined level plus

the standard base sidetone loss.

At block 37, depending on the result of the previous operations, if the

target sidetone loss is less than the current sidetone loss, then the sidetone loss

is reduced at a predetermined rate as determined by the decay timer, block

38. If the target sidetone loss is greater than the current sidetone loss, then the

sidetone loss is increased at a predetermined rate as determined by the attack

timer, block 39. The sequence of blocks 38-40-41/43-45-46/48 shows the sidetone loss

decay or sidetone loss removal process. The decay timer is decremented in 38

by one when the LST algorithm is in the decay process. If the decay timer has

not expired in block 40, the algorithm is exited in box 41. When the decay

timer has expired, the current sidetone value is reduced by one step and the

decay timer is re-initialized as shown in box 43. In box 45, the current

sidetone setting is then compared to the target setting. If the current setting is

less than the target setting, the current setting is made equal to the target

setting in box 46 which means the sidetone has now stabilized. If the current

setting is not less than the target, then the LST algorithm is exited in box 48

and the decay process will continue on the next iteration of the algorithm until

the sidetone loss has reached the new value.

The sequence of blocks 39-42-41/44-47-46/48 shows the sidetone loss

attack or sidetone loss insertion process. The attack timer is decremented in

39 by one when the LST algorithm is in the attack process. If the attack timer

has not expired, the algorithm is exited in box 41. When the attack timer has

expired, the current sidetone value is increased by one step and the attack

timer is re-initialized as shown in box 44. In box 47, the current sidetone

setting is then compared to the target setting. If the current setting is greater

than the target setting, the current setting is made equal to the target setting in

box 46 which means the sidetone has now stabilized. If the current setting is not greater than the target, then the LST algorithm is exited in 48 and the

attack process will continue until the sidetone loss has reached the new value.

The sidetone loss characteristic versus the microphone noise level and

the rate of insertion and removal of sidetone loss is based on subjective

preference and can be determined by means known to those skilled in the art.

As an example, one possible listener sidetone reduction characteristic versus

room noise may be a linear change in sidetone versus room noise i.e. for every

one decibel change in room noise, the sidetone is changed by the same amount.

In one embodiment of the invention, the method described herein is

used to adjust listener sidetone at a wireline telephone handset such as used

in a pay telephone or in a wireless set. However, the method can also be

applied to headsets, whereby the use wears a microphone/ speaker device

over their head. Similarly, the invention can be implemented via a line

interface card or other network device as opposed to the user's terminal. In

this case, the line interface card would adjust the listener sidetone perceived

by the end user.

Claims

WHAT IS CLAIMED IS:

1. A method of adjusting listener sidetone comprising the steps of:

measuring ambient noise of a received input signal;

comparing the measured ambient noise level against a

predetermined threshold; and

adjusting the sidetone level by a predetermined amount in

response to whether the measured ambient noise level is higher or

lower than said predetermined threshold.

2. A method as defined in claim 1, wherein said ambient noise level is measured by discriminating noise from speech received at said input

signal.

3. A method as defined in claim 2, wherein said ambient noise

level is measured using a noise estimator.

4. A method as defined in claim 3, wherein said noise estimator

discriminates noise from speech by rectifying said input signal followed by a

slow attack and fast decay processing of the signal.

5. A method as defined in claim 1, wherein if said ambient noise

level is less than said predetermined threshold, a target sidetone loss is

computed to remain at a standard base level.

6. A method as defined in claim 5, wherein if said ambient noise

level is greater than said predetermined threshold, a target sidetone loss is

computed according to a predetermined level plus the standard base sidetone

loss.

7. A method as defined in claim 6, wherein if said target sidetone

loss is less than a current sidetone loss, then the sidetone loss is reduced at a

predetermined rate as determined by a decay timer.

8. A method as defined in claim 6, wherein if said target sidetone

loss is greater than a current sidetone loss, then the sidetone loss is increased

at a predetermined rate as determined by an attack timer.

9. A method as defined in claim 1, wherein said input signal is

received at a handset microphone.

10. A method as defined in claim 9, wherein said handset is a

wireline handset.

11. A method as defined in claim 9, wherein said handset is a

wireless handset.

12. A method as defined in claim 1, wherein said input signal is

received at a headset microphone.

13. A method as defined in claim 1, wherein said input signal is

received at a network device.

14. A method as defined in claim 13, wherein said network device is

a line interface card.

15. An apparatus for adjusting listener sidetone, comprising:

a noise estimator for measuring ambient noise level received

from an input signal;

means for comparing the measured ambient noise level against a

predetermined threshold; and

a listener sidetone controller for adjusting the sidetone level by a

predetermined amount in response to whether the measured ambient noise

level is higher or lower than said predetermined threshold.

16. An apparatus as defined in claim 15, wherein if said ambient

noise level is less than said predetermined threshold, a target sidetone loss is

computed to maintain a standard base level.

17. An apparatus as defined in claim 16, wherein if said ambient

noise level is greater than said predetermined threshold, a target sidetone loss

is computed according to a predetermined level plus the standard base

sidetone loss.

18. An apparatus as defined in claim 17, wherein if said target

sidetone loss is less than a current sidetone loss, then the sidetone loss is

reduced at a predetermined rate as determined by a decay timer.

19. An apparatus as defined in claim 17, wherein if said target

sidetone loss is greater than a current sidetone loss, then the sidetone loss is

increased at a predetermined rate as determined by an attack timer.

20. An apparatus as defined in claim 15, wherein said input signal is

received at a handset microphone.

21. An apparatus as defined in claim 1, wherein said input signal is

received at a handset microphone.

22. An apparatus as defined in claim 21, wherein said handset is a

wireline handset.

23. An apparatus as defined in claim 21, wherein said handset is a

wireless handset.

24. An apparatus as defined in claim 15, wherein said input signal is

received at a headset microphone.

25. An apparatus as defined in claim 15, wherein said input signal is

received at a network device.

26. An apparatus as defined in claim 25, wherein said network

device is a line interface card.