JPH051168Y2 - - Google Patents

Description

[Detailed explanation of the invention] [Summary] In hands-free telephones that use a microphone and speaker, howling during calls can be reduced without increasing the amount of hardware by averaging the levels of the transmitting and receiving signals using software. prevent.

[Industrial application field]

The present invention relates to a hands-free telephone that uses software to prevent howling during a call.

Hands-free phones that use a microphone and speaker do not require you to hold a handset, so you can safely talk while driving. However, since the volume that circulates from the speaker to the microphone is large, countermeasures against howling are essential.

[Conventional technology]

A hands-free phone call is no different from a normal handset phone call, but feedback is likely to occur because there is a large amount of feedback from the speaker to the microphone.
FIG. 6 is an explanatory diagram of this.

In the figure, 5 is a microphone, 6 is a speaker, and 7 is a microphone.
is a transmitting amplifier, 8 is a receiving amplifier, and 9 is a side sound circuit (two-wire four-wire exchanger). The transmission signal that has been acousto-electrically converted by the microphone 5 is amplified by the transmission amplifier 7 and sent to the line through the side sound protection circuit 9. On the contrary, the receiving signal sent through the line enters the receiving amplifier 8 through the side sound circuit 9,
Here, the signal is amplified and subjected to electro-acoustic conversion by the speaker 6.

Howling in this type of hands-free telephone is caused by sidetone leaking from the transmitting amplifier 7 to the receiving amplifier 8 due to imperfections in the side sound protection circuit 9, return from the other party's side sound circuit, and even from the receiver of the other party. This is due to the line return such as sound going around to the transmitter,
Above all, there is a problem in that the acoustic coupling between the speaker 6 and the microphone 5 forms a closed loop.

Therefore, conventionally, the gain of the receiving amplifier 8 is lowered to keep the loop gain to less than 1 during the period when the transmitting signal is being detected, and the gain of the transmitting amplifier 7 is lowered during the period when the receiving signal is being detected. The gain is kept below 1 to prevent howling.

[The problem that the idea aims to solve]

However, if the signal level is detected using simple hardware, malfunctions are likely to occur. For example, when a peak detection type level detection circuit detects the signal level, if the level of the transmitted signal suddenly increases suddenly while receiving a call, the received sound may be interrupted, and as a result, the transmitted signal level may fluctuate. occurs. However, a hardware configuration such as a smoothing circuit to improve this point requires the addition of one or more ICs, resulting in an increase in the size of the device configuration.

The present invention attempts to eliminate the influence of sudden level fluctuations on gain control by averaging signal levels using software.

[Means to solve the problem]

FIG. 1 is a basic configuration diagram of the present invention, in which 1 is an A/D converter that samples each voice signal of the transmitter and receiver at fixed intervals and converts it into digital data; 3 is the variable gain circuit on the transmitter side; 4 is a variable gain circuit on the transmitting side, and 2 is a processor (CPU) that controls these gains.

[Effect]

The CPU 2 periodically takes in the digital data of the received signal and the transmitted signal via the A/D converter 1, averages them (for example, arithmetic average of N times), and levels the average received signal level and the average transmitted signal level. is constantly calculated. When the average received level is high, it is determined that the voice is being received, and the gain of the variable gain circuit 3 on the transmitting side is lowered, and conversely, when the average transmitted level is high, it is determined that the voice is being transmitted, and the variable gain on the receiving side is reduced. The gain of the circuit 4 is lowered (the gain of the variable gain circuit 3 on the transmitting side is returned).

Since the averaged transmission or reception level does not reflect sudden fluctuations in the audio signal so much, it is possible to avoid interruptions in the reception sound and fluctuations in the level of the transmission signal due to the operation of the variable gain circuit.

〔Example〕

FIG. 2 is a block diagram showing an embodiment of the present invention, and the parts within the broken line frame are the same as those in FIG. 1. 5 is the transmission tone T-A
6 is the receiving sound R-A.
7A and 7B are fixed gain amplifiers on the transmitting side, and 8A and 8B are fixed gain amplifiers on the receiving side. A car phone requires a wireless device T/U, so amplifiers 8A and 7B are connected to this wireless device T/U.
Connect to the side.

In a hands-free telephone, when transmitting a call, the gain of the variable gain circuit 4 on the receiving side is lowered so that the receiving sound R-A is transmitted to the microphone 5.
Furthermore, when receiving a call, the gain of the variable gain circuit 3 on the transmitting side is lowered to reduce the amount of the transmitted speech sound TA (especially noise) picked up by the microphone 5 reaching the other party. Therefore, CPU2 uses A/D converter 1
is used to constantly monitor the output levels of amplifiers 7A and 8A during a call.

FIG. 3 shows a specific example, in which the CPU 2 includes an input/output interface 21, a microprocessor (MPU) 22, a program ROM (read-only memory) 23, and a random access memory 24. 25 is a key switch for entering a telephone number;
26 is a number display for displaying a telephone number. The car phone exchanges various signals with the wireless device T/U.

FIG. 4 is an explanatory diagram of the averaging process of audio signals, and a
b is the case where the transmitted voice is input as A/D input, and b is the case where the received voice is input as A/D. The transmission sound of a is
Sampled every 0.4ms. When the A/D data of each sampling point is T・An, that 16
The first average value T.Am is obtained by arithmetic averaging the batches. This processing time takes approximately 6ms.

Furthermore, this primary average value T.Am is arithmetic averaged for 16 times to obtain a secondary average value T.A _NEW . This processing time is 6×16≈100ms.

Then, compare the current secondary average value T・A _NEW with the previous average level T・A, and calculate T・A _NEW T・A
(when the transmitting volume increases), T・A _NEW is set as a new T・A.

T.A=T.A _NEW ...On the other hand, when T.A _NEW <T.A (when the transmission volume decreases), T.A-1 is set as the new T.A.

T.A=T.A-1... On the other hand, the received sound of b is also sampled every 0.4 ms, but the method of determining the average level R.A is different. The A/D data of each sampling point of the received voice is R.
When An is used, the method for determining the primary average value R·Am and the secondary average value R·A _NEW is the same as for the outgoing voice.

However, in comparison with the previous average level R・A, when it becomes R・A _NEW R・A (when the listening volume increases), it changes as R・A=R・A _NEW +R・A/2... It differs from the outgoing tone in that it has a gentler tone. When R・A _NEW <R・A (when the listening volume decreases), R・A=R・A−1 .

Since the formula for increasing the listening volume is affected by the previous average level, the change is more gradual than the formula for increasing the transmitting volume. This is a transmitting priority mode in which the gain of the electronic volume 3 on the receiving side is lowered when the average receiving level R.A becomes the same as the average transmitting level T.A on the line return, and the transmitting priority mode lowers the gain of the electronic volume 4 on the transmitting side. This is to prevent hunting, which occurs when switching frequently between the receive priority mode and the receive priority mode, which lowers the priority, and to place more weight on the transmit priority mode.

FIG. 5 is a flowchart showing this averaging process. Step S1 is a mode determination process immediately after the line is connected, and in this example, it starts with the transmission sound priority mode. In this mode, the attenuation amount of the electronic volume 3 on the transmitting side is set to, for example, 0 dB, while the attenuation amount of the electronic volume 4 on the receiving side is set to 12 dB to mute the receiving sound. Incidentally, the received voice priority mode, which will be described later, reverses this relationship of attenuation amount and mutes the transmitted voice sound.

Steps S2 to S9 are processes for calculating the average value of the transmitted and received sounds. That is, in steps S3 to S6, 16 transmission sounds (T・An) from n=0 to n=15
and received voice (R・An) every 0.4ms, and based on that, in step S7, the linear average value T・An of the formula is calculated.
Calculate Am, R・Am. From m=0 to m
After repeating until =15 (step S8), step
In S9, the quadratic average value of the formula T・A _NEW , R・A _NEW
Calculate.

Steps S10 to S15 calculate the obtained quadratic average value T・
This is a process of determining the average levels T.A. and R.A. of the transmitted and received sounds based on _{A.sub.NEW and} R.A.NEW. That is, in step S10, T・A _NEW is compared with the previous T・A, and if T・A _NEW T・A, then in step S11, T・A=T・A _NEW (formula), and T・A _NEW <
If T・A, then in step S12 T・A=T・A−
1 (formula). On the other hand, in step S13, R.
Compare A _NEW and R・A, and if it is R・A _NEW R・A, in step S14, R・A=(R・A _NEW +R・A
A)/2 (formula), and if R.A _NEW <R.A, then in step S15 R.A=R.A-1 is set (formula).

Steps S16 to S19 are processes for comparing T.A and R.A to determine the amount of attenuation of the transmitted and received sounds. That is, in step S16, T.A. and R.A. are compared, and if T.AR.A, the outgoing tone is set to priority in step S17, and if T.A.<R.A, the receiving tone is set in step S18. Set sound priority. In step S19, according to the determination of these priority modes, the attenuation amount (0 dB in the above example) is set for electronic volume 3 and 4.
12dB) setting data.

Note that the method for determining the average levels T·A and R·A is not limited to the above example. For example, a weighted average method such as T.A=15/16T.A+1/16T.An R.A=15/16R.A+1/16R.An may be used. Further, in this case, the rate of change in T·A and R·A can be adjusted by each coefficient value.

[Effect of idea]

As described above, according to the present invention, gain (attenuation) control to prevent howling during hands-free phone calls is performed by detecting the respective average levels of the transmitting sound and the receiving sound. This has the advantage of not being affected by changes in volume.

[Brief explanation of the drawing]

Fig. 1 is a basic block diagram of the present invention, Fig. 2 is a block diagram of an embodiment of the present invention, Fig. 3 is a block diagram of a specific example of the present invention, and Fig. 4 is an example of the averaging process of the present invention. FIG. 5 is a flowchart showing the overall processing, and FIG. 6 is a configuration diagram of a conventional hands-free telephone.

Claims (1)

[Scope of Claim for Utility Model Registration] In a hands-free telephone set in which a microphone 5 is used as a transmitter and a speaker 6 is used as a receiver, an A/D converter 1 converts a transmitting signal and a receiving signal.
The data is converted into digital data and taken into the processor 2, and the data is averaged by the processor 2 to obtain the average level of the transmitting and receiving calls, and the gain of the receiving system during the call and the transmitting system are determined according to the comparison results between the two. A hands-free telephone characterized by controlling one of the talking system gains to be high and the other to be low.