JPS61121624A - Echo control system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an echo control method in an electroacoustic system used, for example, in a conference system.

When a speaker and microphone are used in combination, such as in a conference system, the sound emitted from the loudspeaker travels through the room's acoustic system to the microphone, often causing howling, which significantly degrades call quality.

<Basically, it makes the conference participants feel uncomfortable and makes the content of the call unclear.

Therefore, it is desired to provide an echo control method that effectively cancels echoes transmitted through such spatial acoustic system paths.

[Conventional technology]

A conventional example will be explained as one of the means for preventing echo.

ll11! :3@ is a block circuit diagram of an echo cancellation system. In the figure, 1 is the self-side microhoy,
2a is the speaker on the other side, 3 is the speaker on the own side, 4 is the microphone on the other side, 6 and 7 are the disconnection switches for the communication bus provided in the transmitting system and the receiving system, and 5 is the output level of the microphone 1 and the input level of the speaker 30. Comparison circuit 0 Conventionally, the electrical sound level output to the speaker 3 and the electrical signal level also input to the microphone 1 are compared in a total comparison circuit 5, and when the former level is high, the signal from the other side (microphone 4) is compared. When it is determined that the voice is to be heard, the path of the transmission system for the output of the comparison circuit 5 is cut off by the switch 6, so that the signal from the microphone 1 does not reach the speaker 2 of the other party.

If the latter level is high, the user determines that it is time for him to speak, and opens the receiving system switch 7 to stop the sound from the speaker. By preventing the formation of an echo bus in this manner, the generation of echoes is prevented.

[Problem that the invention seeks to solve]

The conventional methods mentioned above have the following drawbacks. In other words, even if level detection is performed in the transmitting direction and in the receiving direction, the boundary between the levels is not clear and it is difficult to clearly distinguish between echoes and actual vocalizations. There is also the problem that the detection delay causes the beginning of the conversation to be cut off, degrading the quality of the call.

As a means of dealing with this, it is possible to apply an echo canceller used in telephone communication systems. However, there are problems in this case as well.

On telephone lines, audio signals are only transmitted in the 4-band frequency band, 0.3 to 3.4i, but in conference systems, etc., the bandwidth is widened to include higher frequencies to improve sound quality. It typically has a bandwidth of 8kHz. In addition, in telephone lines, the echo bus consists only of electrical circuits, but in conference systems, etc., the echo bus section is the section where voice propagates through space, so the echo delay amount is slow and large.◇Signal bandwidth If the width is wide and the amount of delay is large, the amount of calculation that the echo canceller must perform becomes enormous, and the echo canceller 1 cannot be used with conventional technology.
It is difficult to achieve this goal.

[Means for solving problems]

The above problem is solved by dividing the transmission signal into low-frequency signal components and high-frequency signal components, and for the low-frequency signal components, the echo components are canceled using an echo canceller whose main component is an acyclic filter. For high-frequency signal components, the echo control method of the present invention is configured such that frequencies are extracted by a comb-shaped filter so that different frequencies are obtained for the audio output of the other speaker and the audio output of the local speaker. Therefore, it is solved.

[Effect]

According to the present invention, an echo canceler sufficiently cancels the low-frequency signal component where the echo components of the acoustic system are concentrated, and a comb-like frequency extraction is performed for the high-frequency signal component, so that both the self-side and the other-side By making the audio outputs of the speakers have the same frequency, the audio from the local speaker is transmitted to the microphone. Even if the signal is dropped, it will not reach the other party's loudspeaker and will be cut off midway, making it possible to prevent echoes from occurring even in high-frequency components.

Since the signal components are divided into high and low frequencies, the frequency bandwidth that the echo canceler must handle does not increase, and the amount of calculation for the echo canceler is halved compared to when it is performed over the entire band, making it easy to implement and reducing the circuit size. It can also be small.

〔Example〕

The gist of the present invention will be explained in detail with reference to illustrated embodiments.0 Figure 1 is a block circuit diagram of an echo control system.O The transmission system includes a microphone 1, a band separation filter 8, an echo canceller 9, a comb filter 10, and an addition filter. 11
The incoming system consists of a band separation filter 12, a comb filter 13, an addition filter 14, and a speaker 3.The incoming signal is converted into a digital signal by an analog-to-digital converter, and then passed through the band separation filter 12 of the incoming system. The signal is separated into two, a high-frequency signal and a low-frequency signal. This can be achieved, for example, with orthogonal mirror filters. The separated high-frequency components are passed through a comb filter 13 that extracts frequency components at regular intervals.
Spectral thinning is performed by This thinned-out signal is added to the separated low-frequency component in an addition filter 14 to drive the speaker 3.

The sound emitted from the speakers at and ＼ propagates through the space in the conference room and goes around to microphone 1. The sound signal from microphone 1 is separated into high-frequency signals and low-frequency signals by band separation filter 8, as in the incoming call system. be done. The wraparound component, that is, the echo component, is also separated into high and low frequencies at this time, and for the high frequency component, a comb filter 10 is used to extract frequency components that are blocked by the comb filter 13 of the receiving system. As a result, echoes are blocked in frequency on the high frequency side.

For low-frequency components, echo cancellation is performed using a known echo canceller 9.

2 passed through an echo canceller 9 and a comb filter 10
The two signals are combined by an addition filter 11 and sent to the line.

FIGS. 2(&) and 29(a) show other embodiments of the transmitting system according to the present invention in block circuit diagrams. The same symbols as in FIG. 1 indicate the same circuits.

Figure 2 (ml, 伽) shows the arrangement of comb filters in the transmission system.
0 and 10' are provided in FIG. 2(a).

Here, the comb-shaped filter 10 shown by the broken line on the own side can be omitted, and the comb-shaped filter 10' can be provided only on the other side.

Figure 2 tb> shows a connection in which only one 10ft comb filter is provided on its own side. In this case, the circuit at the other end is simplified, and band separation and synthesis circuits can also be omitted.

〔Effect of the invention〕

As described above, according to the present invention, the delay time is large;
The echo control method, which was difficult to implement for audio signals with a relatively wide bandwidth, can now be easily implemented, and can be provided with a simple circuit without complicating the circuit configuration. The effect is remarkable.

FIG. 2(a) is a block circuit diagram of a comb filter arrangement and transmission system configuration in an embodiment of the present invention, and FIG. 3 is a block circuit diagram of a known echo cancellation method.

In the figure, 1.4 is a microphone, 2.3 is the speaker, 5 is a voltage comparison circuit; 6.7 is a switch circuit, 11, 11', and 14 indicate addition filters.

Fu ((2) γ Kaya 2nd day speed h 3 Figure

Claims (1)

[Claims] Divide the transmission system signal into a low-frequency signal component and a high-frequency signal component,
For low-frequency signal components, echo components are canceled using an echo canceller whose main component is an acyclic filter.
The echo control method is characterized in that, for high-frequency signal components, frequency extraction is performed using a comb filter so that the audio output from the other speaker and the audio output from the own speaker have different frequencies.