KR101071895B1 - Adaptive Sound Generator based on an Audience Position Tracking Technique - Google Patents

Abstract

The present invention relates to an adaptive sound generator using a listener position tracking technique that uses listener position information for sound compensation in addition to the head transfer function, and is characterized in that it extends the range of the optimal listening position of the listener.
Further, when estimating the listener position information, the energy detector can be used to acquire the listener position information resistant to noise characteristics.

Description

Adaptive Sound Generator based on an Audience Position Tracking Technique

The present invention relates to an adaptive sound generator using a listener position tracking technique, and more particularly, to an adaptive sound generator capable of realizing an effect of widening an optimal listening range by including a listener position tracker and a sound compensator.

Recently, multimedia technology has developed from a simple sound transmission to a technology that delivers real-world effects. In the music that plays only a rhythm, research and development of sound technology that can create a more grand and realistic atmosphere with various sound effects, and moreover, sound that is close to the original sound is progressing. Among these technologies, stereophonic sound is the most actively studied technology.

As shown in FIG. 1, an optimal listening position of stereoscopic sound is a position of a vertex remaining when there is a speaker on one side of left and right sides of an equilateral triangle. It is known that when the listener is in this position, the best sounding, sound field control, mutual interference cancellation, and virtual surround effect can be felt. In FIG. 1, a thick circle surrounding the listener is a sweet-spot, a virtual sphere in which various sound effects can be felt in the same manner, and is called an equalization zone. As shown in FIG. 1, the optimal listening position of the listener corresponds to a range of circles within a radius of 0.25 m from P ⁰ , but the listener may be located anywhere in the effective space of the speaker, thereby changing the sense of space of the sound source.

Thus, in an adaptive sound generator that automatically adjusts audio reproduction based on the specific audio characteristics of the surrounding environment, compensation is necessary to widen the listener's optimal listening position.

One such compensation method is the Head-Related Transfer Function (HRTF) method. The head transfer function is a function of modeling a path between a sound source and an ear in free space and is an important acoustic parameter related to sound localization. To obtain the head transfer function, the dummy header microphone modeled by the auditory organ is used to obtain the impulse response of both the left and right for each position with a specific distance between the 360 degree azimuth angle and the 180 degree altitude angle. The head transfer function measures a dummy header at a constant position without moving, and loses its characteristics where the listener is moving.

Therefore, there is a need for the realization of additional sound compensation by information other than the head transfer function in the adaptive sound generator for realizing better stereo sound.

SUMMARY OF THE INVENTION The present invention has an object of solving the above-described technical problem, and implements an adaptive sound generator by a listener position tracking technique using the listener's position information for sound compensation in addition to the head transfer function. The aim is to produce more high quality stereo sound by widening the range.

An adaptive sound generator according to an embodiment of the present invention includes a listener location tracker and a sound compensator. The listener position tracking unit includes a multi-channel microphone for receiving a sound signal generated from the listener, a sound detector for detecting the sound signal input from the microphone, and a position estimator for calculating the position information of the listener by the detected sound signal.

In addition, the sound detector is an analog digital converter for converting a sound signal into a digital signal in order to process the input sound signal, and receives the initial signal of a predetermined time to calculate the energy level to derive the noise characteristics of the current environment, It includes an energy detector that outputs only the value above the threshold by setting the threshold based on the noise characteristics.

The sound compensator includes a head transfer function database, and compensates for the sense of space received from a separate device by using the head transfer function corresponding to the listener position from the listener position information by the listener position tracking unit.

According to the adaptive sound generator based on the listener position tracking technique of the present invention, the stereoscopic sound can be generated by widening the range of the optimal listening position by using the position information of the listener in addition to the head transfer function for sound compensation.

Further, by using an energy detector for estimating listener position information, listener position information that is resistant to noise characteristics can be obtained.

Figure 1 shows the optimal position of the listener when applying the image positioning, sound field control, mutual interference cancellation technology.
2 is a block diagram illustrating a configuration of an adaptive sound generator according to an embodiment of the present invention.
3 is a detailed block diagram of the sound detector in the block diagram of FIG. 2.
4 shows a head transfer function in a stereo optimal sound listening environment.
5 is the location of the listener in a normal listening environment.

Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the adaptive sound generator by the listener position tracking technique according to an embodiment of the present invention. The following examples of the present invention are intended to embody the present invention, but not to limit or limit the scope of the present invention. It will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

First, FIG. 2 is a block diagram of an embodiment of an adaptive sound generator. As shown in FIG. 2, the adaptive sound generator based on the listener location tracking method according to an embodiment of the present invention is a listener location tracker that obtains the location information of the listener by sound signals generated from the listener by the multichannel microphone. 100 and a general sound (acoustic sound) input from a separate device as a sound compensated for the sense of space according to the position of the listener using the listener position tracking information and the head transfer function by the listener position tracker. It includes a sound compensator 200 to generate. Specifically, the listener position tracking unit 100 is a multi-channel microphone 10 for receiving a sound signal generated from the listener, a sound detector 20 for detecting a sound signal input from the multi-channel microphone 10 and the detected sound signal It includes a position estimator 30 for calculating the position information of the listener.

3 illustrates, as an embodiment of the sound detector 20, an analog digital converter 21 and an energy detector 22 for converting an input sound signal into a digital signal.

Specific operation of the energy detector 22 will be described below.

An initial signal of a certain time is received from the digitally converted sound signal, the noise characteristic of the current environment is derived using the initial signal, and a threshold value is set based on the noise characteristic. When the threshold setting is completed, the sound signal to be used for tracking the location information is acquired, and the energy level of the acquired signal is calculated. If the calculated energy level is higher than the threshold value, the position is calculated. Information is sent to the estimator, otherwise the sound signal is acquired again. This operation of the energy detector 22 may enhance the noise characteristics of the sound signal to be used to obtain listener position estimate information in the position estimator 30.

A preferred embodiment of the operation of the position estimator 30 will be described below.

The position estimator 30 is characterized by obtaining final position information of the listener by obtaining a time difference of arrival of a sound signal input (TDoA positioning method). The TDoA positioning method does not measure absolute visual information but measures only the time difference received, so it is easier to implement than the ToA (Time of Arrival) positioning method, which measures the distance of sound source transmission between the sound source and the receiver to find the distance. There is an advantage.

Gauss-Newton-based least-squares method is a method of obtaining final position information of the listener from the time difference of arrival of the sound signal obtained by the TDoA positioning method. .

In Equation 1, A is the difference between the position of the current multi-channel microphone 10 and the initial position value, δ is the error of the position value, m is the position value of the listener actually calculated by the position estimator 30, f is the actual listener Where z denotes the difference between the position value m of the listener actually calculated by the position estimator 30 and the position f of the actual listener.

A detailed operation of the sound compensator 200 will be described below.

The sound compensator 200 has a built-in database of a head transfer function for compensating according to the pre-measured listener position, receives the listener position information from the listener position tracker 100, and receives The head transfer function corresponding to the position of the listener is obtained from the database, and the inverse filtering is performed using the acquired head transfer function to compensate for the acoustic space sense according to the position of the listener.

A detailed mathematical approach for an embodiment in which the sound compensator 200 performs inverse filtering using a head transfer function using listener position information is as follows.

4 shows a head transfer function in a stereo optimal sound listening environment. It is assumed that the positions of the two speakers and the listener are in a plane parallel to the ground. That is, the elevation angle is 0 degrees, and the head transfer function is designed by considering only the azimuth angle.

는 최적 청취 환경에서의 청취자의 위치,

과

은 청취자와 좌우 스피커를 잇는 직선을 향한 법선을 기준으로 좌우 스피커와

를 잇는 직선의 사잇각이다.In Figure 4

Is the position of the listener in the optimal listening environment,

and

The left and right speakers are based on a normal line facing the listener and the left and right speakers.

The angle of the straight line connecting the.

이며, 최적 청취 위치에서의

과

은 일반 청취 위치에서

과

으로 사잇각을 정의한다. 왼쪽과 오른쪽 귀에 수신되는 음향신호

와

는 수학식 2와 수학식 3과 같다.The position of the listener in the general listening environment as shown in FIG.

At the optimal listening position

and

In the normal listening position

and

To define the angle. Acoustic signal received at left and right ears

Wow

Is the same as Equation 2 and Equation 3.

과

는

방향특성을 갖는 머리전달함수,

와

는 좌우 스피커로 송출하는 음향 신호,

는 전파 지연이다. 수학 식2와 수학식 3에서

과

는 혼성음이다. 수학식 2와 수학식 3을 행렬식을 사용해 간단히 수학식 4처럼 다시 쓸 수 있다.here

and

Is

Head transfer function with directional characteristics,

Wow

Sound signal transmitted to left and right speakers,

Is the propagation delay. In Equation 2 and Equation 3

and

Is a mixed sound. Equations 2 and 3 can be rewritten as equation 4 simply using determinants.

,

,

here ,

,

,

는 전치 행렬을 의미한다. 수식 표현의 용이성을 위해

는 표현하지 않는다. ego,

Is the transpose matrix. For ease of expression

Does not express.

 The optimal acoustic signal to which the head transfer function is applied can be defined as in Equation 5.

이다.here

to be.

과

)를 추정하여 청취자의 위치와 무관하게 정확한 입체 음향을 제공한다. 본 발명의 일실시예에서는 수학식 6처럼 머리전달함수를 이용한 최적 입체 음향 시스템을 설계한다In order to make Equation 4 like Equation 5, spatial information (especially

and

) To provide accurate stereo sound regardless of the listener's position. In an embodiment of the present invention, the optimal stereo sound system is designed using the head transfer function as shown in Equation (6).

100: listener position tracking unit 200: sound compensation unit
10: multichannel microphone 20: sound detector
30: position estimator
21: analog digital converter 22: energy detector

Claims (9)

A listener position tracking unit configured to receive a sound signal generated from the listener and obtain position information of the listener from the received sound signal; And
A sound compensator configured to generate a sound compensated for the sense of space according to the position of the listener by compensating the sense of space of the general sound by using the listener position information obtained through the listener position tracker; Including,
The listener position tracking unit,
A multi-channel microphone for receiving a sound signal generated from the listener;
A sound detector for detecting a sound signal received by the multichannel microphone; And
A position estimator for calculating position information of the listener based on the detected sound signal; Including;
The sound detector,
An analog digital converter for converting an input sound signal into a digital signal; And
Receive the initial signal from the digitally converted sound signal and calculate the energy level to derive the noise characteristics of the current environment, and set the threshold value based on the derived noise characteristics, so that the energy level of the sound signal is above the threshold An energy detector for outputting only values; Adaptive sound generator according to the listener position tracking technique comprising a.
The method of claim 1, wherein the sound compensator,
A head transfer function database is built-in, and the head position transfer function corresponding to the position of the listener is acquired from the database from the listener position tracking unit, and the inverse filtering using the acquired head transfer function is performed to sense the space according to the position of the listener. Adaptive sound generator by listener position tracking technique characterized in that performing the compensation.
3. The adaptive sound generator of claim 2, wherein the inverse filtering using the head transfer function is performed by the following equation.

(In the above equation,

Is the acoustic signal compensated by the sound compensator,

Is the inverse of the head transfer function,

Is the optimal acoustic signal with head transfer function.)
delete delete The method of claim 1,
The position estimator uses Gauss-Newton-based least squares method to calculate the difference of arrival time of the sound signal by TDoA positioning function and to obtain the final position information of the listener using the calculated difference of arrival time. Adaptive sound generator according to the listener position tracking technique.
The method of claim 6,
Gauss-Newton-based least-squares method is an adaptive sound generator based on the listener position tracking method characterized by the following equation.

Where A is the difference between the position of the current multichannel microphone and the initial position value, δ is the error of the position value, m is the position value of the listener actually calculated by the position estimator, and f is the position of the actual listener. Where z is the difference between the position value m of the listener actually calculated by the position estimator and the position f of the actual listener.)
delete delete

Images