JPH0572025A - Estimating method for position of sound source - Google Patents

Estimating method for position of sound source

Info

Publication number
JPH0572025A
JPH0572025A JP3230139A JP23013991A JPH0572025A JP H0572025 A JPH0572025 A JP H0572025A JP 3230139 A JP3230139 A JP 3230139A JP 23013991 A JP23013991 A JP 23013991A JP H0572025 A JPH0572025 A JP H0572025A
Authority
JP
Japan
Prior art keywords
sound source
source position
sound
virtual
source
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP3230139A
Other languages
Japanese (ja)
Other versions
JP2760177B2 (en
Inventor
Takeo Oono
剛男 大野
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP3230139A priority Critical patent/JP2760177B2/en
Publication of JPH0572025A publication Critical patent/JPH0572025A/en
Application granted granted Critical
Publication of JP2760177B2 publication Critical patent/JP2760177B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Landscapes

  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

PURPOSE:To obtain the method for estimating the position of a sound source wherein the increase in the amount of computations is prevented when a space for assuming the presence of the sound source is expanded and the arranging interval of the positions of the virtual sound sources is narrowed so as to enhance the position estimating accuracy. CONSTITUTION:The position of a sound source is estimated by using a plurality of microphones. In estimation, a small number of virtual-sound-source positions (grid points 1) are arranged in a two-dimensional space, wherein the sound source is assumed to be present, and in the same space as the arranging interval of the virtual-sound-source positions required for estimating accuracy at the first estimation of the sound-source position. The sound-source position is estimated among the virtual-sound-source positions. Then, the virtual-sound-source positions are arranged at the interval, which is narrower than that at the first estimation in the space, which is surrounded with the neighboring virtual-sound- source positions (grid points 12a-12h), with sound-source position 11 obtained at the first estimation as the center. The sound-source position is finally estimated among the virtual-sound-source position at this time.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は、複数のマイクロホンを
配置して音源位置を推定する音源位置推定方法に関する
ものである。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sound source position estimating method for estimating a sound source position by arranging a plurality of microphones.

【0002】[0002]

【従来の技術】マイクロホンアレイを用いた音源位置の
推定に際し、音源とマイクロホンの距離が比較的近距離
にある場合、音源から放射される音波を球面波とみなし
て音源位置を推定することが有効であることは文献「多
数センサによる音源位置の推定」日本音響学会誌46巻
7号(1990)に詳しく述べられている。この従来の
音源位置推定方法では、音源が存在すると考えられる空
間内に複数の仮想音源位置を想定し、マイクロホンアレ
イを構成する各マイクロホンより得られる信号のスペク
トルについて先の仮想音源位置と各マイクロホンとの距
離に基づき位相と振幅を補正した後、すべてのマイクロ
ホンにおける補正後のスペクトルを加算平均する。加算
平均された平均音源スペクトルZ(ω,i)は(数1)
のように表わされる。
2. Description of the Related Art When estimating the position of a sound source using a microphone array, it is effective to consider the sound wave emitted from the sound source as a spherical wave and estimate the position of the sound source when the distance between the sound source and the microphone is relatively short. That is described in detail in the document “Estimation of Sound Source Position by Multiple Sensors”, Journal of Acoustical Society of Japan, Vol. 46, No. 7 (1990). In this conventional sound source position estimation method, a plurality of virtual sound source positions are assumed in the space where a sound source is considered to exist, and the virtual sound source position and each microphone are compared with the previous virtual sound source position with respect to the spectrum of the signal obtained from each microphone that configures the microphone array. After the phase and amplitude are corrected based on the distance of, the corrected spectra of all microphones are added and averaged. The averaged sound source spectrum Z (ω, i) obtained by averaging is (Equation 1)
It is expressed as.

【0003】[0003]

【数1】 [Equation 1]

【0004】ここで、Mはマイクロホンアレイを構成す
るマイクロホンの総数、Xm(ω)はm番目のマイクロ
ホンから得られるスペクトルを角周波数ωの関数として
表わしたもの、iはすべての仮想音源位置に順番に付け
た番号、rimはi番目の仮想音源とm番目のマイクロホ
ンとの距離、cは音速である。ただし、各マイクロホン
より得られる信号のS/N比は、マイクロホンと実際の
音源の距離が遠くなるほど低下すると考えられるので、
(数1)における加算平均においては、各項にrimの反
比例項を重み関数として乗じた後、平均を行なってい
る。
Here, M is the total number of microphones forming the microphone array, X m (ω) is the spectrum obtained from the m-th microphone as a function of angular frequency ω, and i is at all virtual sound source positions. The numbers given in order, r im is the distance between the i-th virtual sound source and the m-th microphone, and c is the speed of sound. However, since the S / N ratio of the signal obtained from each microphone is considered to decrease as the distance between the microphone and the actual sound source increases,
In addition averaging in (Equation 1), each term is multiplied by the inverse proportional term of r im as a weighting function, and then the average is performed.

【0005】この平均音源スペクトルのパワー|Z
(ω,i)|は,仮想音源位置と実際の音源位置とが一
致したときに最大値をとり、この平均音源スペクトルの
パワーの最大値をもたらす仮想音源位置を実際の音源位
置と推定する。したがって、音源位置を推定するために
は、実際の音源が存在すると考えられる空間全体にわた
って、平均スペクトルのパワーを仮想音源位置の関数と
して表わす必要がある。
Power of this average sound source spectrum | Z
(Ω, i) | takes a maximum value when the virtual sound source position and the actual sound source position coincide with each other, and estimates the virtual sound source position that gives the maximum value of the power of the average sound source spectrum as the actual sound source position. Therefore, in order to estimate the sound source position, it is necessary to represent the power of the average spectrum as a function of the virtual sound source position over the space where the actual sound source is considered to exist.

【0006】[0006]

【発明が解決しようとする課題】しかしながら、上記従
来の音源位置推定方法では、音源が存在すると考えられ
る空間が広範囲に広がったり、音源位置の推定精度を高
くするために仮想音源位置の間隔を狭めたりする必要が
ある場合には、平均音源スペクトルを求めなければなら
ない仮想音源位置の数が幾何級数的に増加し、これに伴
い、音源位置を推定するための計算量が増加してしまう
という問題を有していた。
However, in the above-mentioned conventional sound source position estimating method, the space in which the sound source is considered to be present spreads over a wide range, and the interval between the virtual sound source positions is narrowed in order to improve the estimation accuracy of the sound source position. However, the number of virtual sound source positions for which the average sound source spectrum must be calculated increases geometrically, and the amount of calculation for estimating the sound source position increases accordingly. Had.

【0007】本発明は、上記従来の問題点を解決するも
のであり、仮想音源位置を想定する空間が拡大された
り、音源位置の推定精度を高める場合にも、計算量の極
端な増加を防止して推定効率を向上させることができる
ようにした音源位置推定方法を提供することを目的とす
る。
The present invention solves the above-mentioned conventional problems, and prevents an excessive increase in the amount of calculation even when the space assuming a virtual sound source position is expanded or the estimation accuracy of the sound source position is improved. It is an object of the present invention to provide a sound source position estimation method capable of improving the estimation efficiency.

【0008】[0008]

【課題を解決するための手段】この目的を達成するため
に本発明による音源位置推定方法は、音源から発する信
号を空間内に配置された複数のマイクロホンで受信し、
この信号のスペクトル成分を各マイクロホンについて求
め、音源が存在すると想定される空間内に仮想した音源
位置とマイクロホンの位置との間の距離に基づき、各マ
イクロホンで受信した信号から求めたスペクトルの位相
と振幅を補正することによって音源のスペクトルを求
め、この音源スペクトルをマイクロホンについての加算
平均した平均音源スペクトルのパワーを仮想音源位置の
関数として計算することにより、真の音源位置を推定す
るに際し、仮想音源の存在想定空間を段階的に絞り込
み、かつ仮想音源位置の配置間隔を段階的に狭めながら
繰り返して音源位置を推定するようにしたものである。
In order to achieve this object, a sound source position estimating method according to the present invention receives a signal emitted from a sound source by a plurality of microphones arranged in space,
The spectral component of this signal is obtained for each microphone, and the phase of the spectrum obtained from the signal received by each microphone is calculated based on the distance between the virtual sound source position and the microphone position in the space where the sound source is assumed to exist. In estimating the true sound source position, the sound source spectrum is obtained by correcting the amplitude, and the power of the average sound source spectrum obtained by averaging the sound source spectrum with respect to the microphone is calculated as a function of the virtual sound source position. Is assumed to be narrowed down in stages, and the sound source position is estimated repeatedly by narrowing the arrangement interval of virtual sound source positions stepwise.

【0009】[0009]

【作用】したがって、本発明によれば、1回の音源位置
推定における仮想音源位置の数を少数にとどめることが
できるので、平均音源スペクトルを求める計算回数を減
少させることができ、音源位置を推定する空間が拡大さ
れたり、仮想音源位置の間隔が狭められたりする場合に
おいても、音源位置を推定するために必要な計算量の極
端な増加を抑えて推定の精度を維持することができる。
Therefore, according to the present invention, since the number of virtual sound source positions in one sound source position estimation can be kept to a small number, the number of calculations for obtaining the average sound source spectrum can be reduced and the sound source position can be estimated. Even when the space to be used is expanded or the interval between virtual sound source positions is narrowed, it is possible to suppress the extreme increase in the amount of calculation required to estimate the sound source position and maintain the estimation accuracy.

【0010】[0010]

【実施例】以下、本発明の一実施例について、図面を参
照しながら説明する。
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

【0011】図2は本発明の一実施例における音源位置
推定方法を説明するため音源とマイクロホンアレイの配
置および仮想音源位置の存在範囲を示す図である。
FIG. 2 is a diagram showing the arrangement of the sound source and the microphone array and the existence range of the virtual sound source position for explaining the sound source position estimating method in one embodiment of the present invention.

【0012】図2において、21は500〜5000H
zの帯域で白色なスペクトルをもつ信号を発する音源、
22a〜22hは音源21の設置された平面から50c
m離れた平面上の半径60cmの円周上にほぼ等間隔に
配置され、円形マイクロホンアレイを構成する8個の無
指向性のマイクロホン、23は音源が存在すると考えら
れ、仮想音源位置を配置する1m四方の2次元平面の範
囲、24はこの2次元空間に導入した直交座標軸x、y
を示す(以下、この座標系の単位はcmとする)。この
座標系において、仮想音源位置を配置する範囲は(数
2)により表わされる。
In FIG. 2, 21 is 500 to 5000H.
a sound source that emits a signal with a white spectrum in the z band,
22a to 22h are 50c from the plane where the sound source 21 is installed
Eight omnidirectional microphones, which are arranged at substantially equal intervals on a circle having a radius of 60 cm on a plane separated by m and form a circular microphone array, are considered to have sound sources, and virtual sound source positions are arranged. A range of a two-dimensional plane of 1 m square, 24 is an orthogonal coordinate axis x, y introduced in this two-dimensional space.
(Hereinafter, the unit of this coordinate system is cm). In this coordinate system, the range in which the virtual sound source position is arranged is represented by (Equation 2).

【0013】[0013]

【数2】 [Equation 2]

【0014】また、実際の音源位置は(0、0)と表わ
される。本実施例においては、各マイクロホン22a〜
22hから得られる信号は、サンプリング周波数12k
Hzで同時サンプリングされた後、512ポイントのサ
ンプリングデータを用いてスペクトル成分Xm(ω)に
変換するものとする。ここで得られたXm(ω)をもと
に、(数1)により平均音源スペクトルZ(ω,i)を
求め、後述のような周波数成分についてこのパワー|Z
(ω,i)|を加算平均し、この最大値を与える仮想音
源位置を推定音源位置とする。本実施例においては、
(数2)の範囲に、x軸方向、Y軸方向ともに2cm間
隔の格子点を想定し、これを仮想音源位置とする。
The actual sound source position is represented as (0,0). In the present embodiment, each microphone 22a-
The signal obtained from 22h has a sampling frequency of 12k.
After being simultaneously sampled at Hz, the sampling data of 512 points is used to convert into the spectral component X m (ω). Based on X m (ω) obtained here, the average sound source spectrum Z (ω, i) is obtained by (Equation 1), and this power | Z
(Ω, i) | is added and averaged, and the virtual sound source position that gives this maximum value is set as the estimated sound source position. In this embodiment,
In the range of (Equation 2), grid points at 2 cm intervals in both the x-axis direction and the Y-axis direction are assumed, and these are set as virtual sound source positions.

【0015】以下、本発明実施例と上記従来例の差異を
理解しやすくするため、両者を対比しながら説明する。
In order to make it easier to understand the difference between the embodiment of the present invention and the above-mentioned conventional example, the two will be described in comparison with each other.

【0016】従来の音源位置推定方法では、図4の仮想
音源位置配置図に示すすべての格子点41について平均
音源スペクトルZ(ω,i)を求める必要がある。本実
施例と同じ状況において、従来の音源位置推定方法によ
って音源位置の推定を行なおうとすると、2601個
(51個×51個)の格子点41について平均音源スペ
クトルZ(ω,i)を求めることになる。図5は上記従
来例のように2cm間隔の2601個の格子点41のす
べてについて、平均音源スペクトルZ(ω,i)を求
め、1000〜1500Hzの周波数成分について加算
平均した結果を示す分布図である。これより音源位置が
(2、0)と推定され、実際の音源位置とよく一致して
いることがわかる。
In the conventional sound source position estimating method, it is necessary to obtain the average sound source spectrum Z (ω, i) for all the grid points 41 shown in the virtual sound source position layout diagram of FIG. In the same situation as the present embodiment, when trying to estimate the sound source position by the conventional sound source position estimating method, the average sound source spectrum Z (ω, i) is obtained for 2601 (51 × 51) grid points 41. It will be. FIG. 5 is a distribution chart showing the results of obtaining the average sound source spectrum Z (ω, i) for all 2601 lattice points 41 at 2 cm intervals as in the above-mentioned conventional example and averaging the frequency components of 1000 to 1500 Hz. is there. From this, it can be seen that the sound source position is estimated to be (2, 0) and is in good agreement with the actual sound source position.

【0017】一方、本発明実施例では、1回の位置推定
で平均音源スペクトルを求める仮想音源位置の数を少数
にとどめ、音源位置の推定を繰り返し、仮想音源位置を
想定する空間を段階的に狭めていく。
On the other hand, in the embodiment of the present invention, the number of virtual sound source positions for which the average sound source spectrum is obtained by one position estimation is limited to a small number, the sound source position estimation is repeated, and the space assuming the virtual sound source position is gradually changed. To narrow.

【0018】図1(a)は本発明実施例における1回目
の音源位置推定の際の仮想音源位置の配置図、図1
(b)は本発明実施例における2回目の音源位置推定の
際の仮想音源位置の配置図である。まず、(数2)の範
囲に図1(a)に示すように、10cm間隔の121
(11個×11個)個の格子点1を想定し、これを初回
の音源位置推定における仮想音源位置とし、加算平均ス
ペクトルZ(ω,i)を求める。これを、500〜10
00Hzの周波数成分について加算平均した結果が図3
(a)であり、これより最大の平均加算スペクトルのパ
ワーを与える点が推定音源位置として(0、0)と求ま
る。次に、図1(b)に示すように、2回目の音源位置
推定のための仮想音源位置の範囲を、初回の音源位置推
定で求めた音源位置11を中心に、1回目の音源位置推
定において隣接した8つの格子点12a〜12hで囲ま
れた空間(数3)となるように設定する。
FIG. 1A is a layout diagram of virtual sound source positions in the first sound source position estimation in the embodiment of the present invention, FIG.
(B) is a layout diagram of virtual sound source positions at the time of second sound source position estimation in the embodiment of the present invention. First, in the range of (Equation 2), as shown in FIG.
Assuming (11 × 11) grid points 1 as virtual sound source positions in the first sound source position estimation, the arithmetic mean spectrum Z (ω, i) is obtained. This is 500-10
Figure 3 shows the result of arithmetic mean for the frequency component of 00Hz.
It is (a), and the point which gives the maximum average addition spectrum power is obtained as (0, 0) as the estimated sound source position. Next, as shown in FIG. 1B, the range of the virtual sound source position for the second sound source position estimation is centered on the sound source position 11 obtained in the first sound source position estimation, and the first sound source position estimation is performed. In (3), the space is set so as to be surrounded by eight adjacent grid points 12a to 12h.

【0019】[0019]

【数3】 [Equation 3]

【0020】この範囲に2cm間隔の121個(11個
×11個)の格子点13を想定し、これを2回目の音源
位置推定における仮想音源位置とし、平均音源スペクト
ルZ(ω,i)を求めた後、2000〜2500Hzの
周波数成分について加算平均すると、図3(b)に示す
ように、スペクトルのパワーの最大値を与える仮想音源
位置が(2、0)と求まり、実際の音源位置とよく一致
していることがわかる。つまり、2回の音源位置推定を
繰り返すことにより、合計242個の仮想音源位置につ
いてのみ平均音源スペクトルを求めるだけで、合計26
01個の仮想音源位置について平均音源スペクトルを求
める場合と同様の結果が得られた。
121 (11 × 11) grid points 13 at 2 cm intervals are assumed in this range, and these are set as virtual sound source positions in the second sound source position estimation, and the average sound source spectrum Z (ω, i) is calculated. After the calculation, the averaging of the frequency components of 2000 to 2500 Hz is performed, and the virtual sound source position giving the maximum value of the spectrum power is calculated as (2, 0) as shown in FIG. You can see that they agree well. That is, by repeating the sound source position estimation twice, the average sound source spectrum is obtained only for a total of 242 virtual sound source positions.
The same result as when the average sound source spectrum was obtained for 01 virtual sound source positions was obtained.

【0021】本実施例においては、2段階の音源位置推
定を行なっているが、この初回の音源位置推定におい
て、500〜1000Hzという、2回目の推定に用い
た周波数帯域よりも低域の周波数帯域を用いているの
は、以下のような理由による。
In this embodiment, the sound source position is estimated in two stages. In this first sound source position estimation, a frequency band of 500 to 1000 Hz, which is lower than the frequency band used for the second estimation, is used. The reason why is used is as follows.

【0022】図3(b)は本実施例より得られた2cm
間隔の仮想音源位置に対する平均音源スペクトルのパワ
ーを2000〜2500Hzという高域の成分について
加算平均した結果である。図5と比較してわかるよう
に、高域のスペクトルを用いた場合には極大値の間隔が
互いに狭まっていることがわかる。したがって、10c
m間隔の格子点1(図1(a)参照)を仮想音源位置と
する初回の音源位置推定において、高域のスペクトルを
用いてしまうと、平均音源スペクトルZ(ω,i)のパ
ワーは、仮想音源位置近傍における実音源の存在の確か
らしさを代表する値とはみなすことができなくなってし
まう。よって、本実施例のように仮想音源位置の間隔を
異ならせて段階的に音源位置推定を行なっていく場合に
は、仮想音源位置の配置間隔に対応して、平均音源スペ
クトルのうち、いずれの帯域を用いるかを決定する必要
がある。図3(b)のように、真の音源位置以外に平均
音源スペクトルのパワーが極大値をもつことは、マイク
ロホン22a〜22h(図2参照)の数や、用いるスペ
クトルの帯域との関係も含めて上記の文献「多数センサ
による音源位置の推定」において考察されている。
FIG. 3 (b) shows the 2 cm obtained from this embodiment.
It is the result of adding and averaging the power of the average sound source spectrum with respect to the virtual sound source positions at intervals for the high frequency components of 2000 to 2500 Hz. As can be seen from comparison with FIG. 5, the intervals between the maximum values are narrowed when the high frequency spectrum is used. Therefore, 10c
When a high-frequency spectrum is used in the first sound source position estimation with the virtual sound source positions at grid points 1 (see FIG. 1A) at m intervals, the power of the average sound source spectrum Z (ω, i) becomes It cannot be regarded as a value representing the certainty of the existence of the real sound source near the virtual sound source position. Therefore, in the case of performing the sound source position estimation stepwise by making the intervals of the virtual sound source positions different as in the present embodiment, one of the average sound source spectra corresponding to the arrangement interval of the virtual sound source positions is selected. It is necessary to decide whether to use the band. As shown in FIG. 3B, the fact that the power of the average sound source spectrum has a maximum value other than the true sound source position includes the relationship between the number of microphones 22a to 22h (see FIG. 2) and the spectrum band to be used. Are considered in the above-mentioned document "Estimation of Sound Source Position by Multiple Sensors".

【0023】なお、本実施例においては、8つのマイク
ロホン22a〜22hを音源21から50cmの高さの
半径60cmの円周上に配置しているが、マイクロホン
アレイを構成するマイクロホンの数およびその配置につ
いては任意に選定することができる。また、本実施例に
おいては、仮想音源の存在想定空間を2次元の平面とし
たが、これを1次元、あるいは3次元の空間としても適
用可能である。
In this embodiment, the eight microphones 22a to 22h are arranged on the circumference having a height of 50 cm from the sound source 21 and a radius of 60 cm. However, the number of microphones constituting the microphone array and the arrangement thereof. Can be arbitrarily selected. Further, in the present embodiment, the assumed existence space of the virtual sound source is a two-dimensional plane, but this may be applied as a one-dimensional or three-dimensional space.

【0024】更に、本実施例では音源位置の推定を2回
繰り返しているが、これを2回以上繰り返すことによ
り、仮想音源の存在想定空間がより広範囲に広がった場
合や、より高精度の推定が望まれる場合に対応すること
が可能である。本発明は、このほか、その基本的技術思
想を逸脱しない範囲で種々変更することができる。
Further, in the present embodiment, the estimation of the sound source position is repeated twice, but by repeating this twice or more, when the assumed space of existence of the virtual sound source is spread over a wider range or the estimation with higher accuracy is performed. Can be dealt with if desired. Besides, the present invention can be variously modified without departing from the basic technical idea thereof.

【0025】[0025]

【発明の効果】以上説明したように本発明によれば、音
源位置推定の計算を段階的に狭められた仮想音源の存在
空間に対して繰り返し行なうことにより、推定の精度を
劣化させることなく、音源位置推定に伴う計算量を削減
させることができ、推定効率を向上させることができ
る。
As described above, according to the present invention, the calculation of the sound source position estimation is repeatedly performed for the existence space of the virtual sound source which is narrowed stepwise, so that the estimation accuracy is not deteriorated. The amount of calculation associated with the sound source position estimation can be reduced, and the estimation efficiency can be improved.

【図面の簡単な説明】[Brief description of drawings]

【図1】(a)本発明の一実施例における音源位置推定
方法に用いる1回目の音源位置推定の際の仮想音源位置
の配置図 (b)同実施例における音源位置推定方法に用いる2回
目の音源位置推定の際の仮想音源位置の配置図
FIG. 1A is a layout diagram of virtual sound source positions at the time of first sound source position estimation used in the sound source position estimation method according to the embodiment of the present invention; and FIG. 1B is a second time of the sound source position estimation method according to the same embodiment. Of virtual sound source position when estimating sound source position

【図2】本発明の一実施例における音源位置推定方法を
説明するための音源とマイクロホンアレイの配置および
仮想音源位置の存在範囲を示す図
FIG. 2 is a diagram showing an arrangement of a sound source and a microphone array and an existing range of a virtual sound source position for explaining a sound source position estimating method according to an embodiment of the present invention.

【図3】(a)本発明の一実施例における音源位置推定
方法により1回目の音源位置を推定する際、500〜1
000Hzの音源成分を用いて音源位置の推定を行なっ
た場合の加算平均スペクトルのパワー分布図 (b)同実施例における音源位置推定方法により2回目
の音源位置を推定する際、2000〜2500Hzの音
源成分を用いて音源位置の推定を行なった場合の加算平
均スペクトルのパワー分布図
FIG. 3A is a diagram illustrating a method of estimating a sound source position for the first time by a sound source position estimating method according to an embodiment of the present invention.
A power distribution diagram of an additive average spectrum when a sound source position is estimated using a sound source component of 000 Hz. (B) When estimating the second sound source position by the sound source position estimating method in the embodiment, a sound source of 2000 to 2500 Hz Power distribution map of the averaging spectrum when the sound source position is estimated using the component

【図4】従来の音源位置推定方法に用いる仮想音源位置
の配置図
FIG. 4 is a layout diagram of virtual sound source positions used in a conventional sound source position estimation method.

【図5】従来の音源位置推定方法による1000〜15
00Hzの音源成分の加算平均スペクトルのパワー分布
FIG. 5: 1000 to 15 according to a conventional sound source position estimation method
Power distribution diagram of averaging spectrum of sound source component of 00Hz

【符号の説明】[Explanation of symbols]

11 1回目の音源位置推定により音源位置と推定され
た点 12 格子点(1回目の音源位置推定により音源位置と
推定された点11に隣接した8つの仮想音源位置) 21 実際の音源 22 マイクロホンアレイを構成する無指向性マイクロ
ホン 23 仮想音源位置の想定する2次元空間の範囲 24 同2次元空間に導入したxy直交座標軸
11 Points estimated as sound source positions by the first sound source position estimation 12 Grid points (8 virtual sound source positions adjacent to the point 11 estimated as the sound source positions by the first sound source position estimation) 21 Actual sound source 22 Microphone array Microphone 23 that composes the range 23 of the assumed two-dimensional space of the virtual sound source position 24 xy orthogonal coordinate axes introduced into the same two-dimensional space

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】音源から発する信号を空間内に配置された
複数のマイクロホンで受信し、この信号のスペクトル成
分を各マイクロホンについて求め、音源が存在すると想
定される空間内に仮想した音源位置とマイクロホンの位
置との間の距離に基づき、各マイクロホンで受信した信
号から求めたスペクトルの位相と振幅を補正することに
よって音源のスペクトルを求め、この音源スペクトルを
マイクロホンについての加算平均した平均音源スペクト
ルのパワーを仮想音源位置の関数として計算することに
より、真の音源位置を推定するに際し、仮想音源の存在
想定空間を段階的に絞り込み、かつ仮想音源位置の配置
間隔を段階的に狭めながら繰り返して音源位置を推定す
ることを特徴とする音源位置推定方法。
Claims: 1. A signal generated from a sound source is received by a plurality of microphones arranged in a space, a spectral component of this signal is obtained for each microphone, and a virtual sound source position and a microphone are assumed in the space where the sound source is assumed to exist. The power source spectrum is calculated by correcting the phase and amplitude of the spectrum obtained from the signal received by each microphone based on the distance between the By estimating as a function of the virtual sound source position, when estimating the true sound source position, the assumed sound space of the virtual sound source is narrowed down step by step, and the sound source position is repeated by gradually narrowing the placement interval of the virtual sound source position. A method for estimating the position of a sound source, which comprises:
JP3230139A 1991-09-10 1991-09-10 Sound source location estimation method Expired - Fee Related JP2760177B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP3230139A JP2760177B2 (en) 1991-09-10 1991-09-10 Sound source location estimation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP3230139A JP2760177B2 (en) 1991-09-10 1991-09-10 Sound source location estimation method

Publications (2)

Publication Number Publication Date
JPH0572025A true JPH0572025A (en) 1993-03-23
JP2760177B2 JP2760177B2 (en) 1998-05-28

Family

ID=16903196

Family Applications (1)

Application Number Title Priority Date Filing Date
JP3230139A Expired - Fee Related JP2760177B2 (en) 1991-09-10 1991-09-10 Sound source location estimation method

Country Status (1)

Country Link
JP (1) JP2760177B2 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5867581A (en) * 1994-10-14 1999-02-02 Matsushita Electric Industrial Co., Ltd. Hearing aid
JP2010139476A (en) * 2008-12-15 2010-06-24 Nittobo Acoustic Engineering Co Ltd Calculation method of sound impedance, and system
JP2013015468A (en) * 2011-07-06 2013-01-24 Hitachi Engineering & Services Co Ltd Abnormal sound diagnostic apparatus and abnormal sound diagnostic method
JP2015059794A (en) * 2013-09-18 2015-03-30 株式会社小野測器 Sound source survey device
US11418871B2 (en) * 2018-05-04 2022-08-16 Sennheiser Electronic Gmbh & Co. Kg Microphone array
CN115150712A (en) * 2022-06-07 2022-10-04 中国第一汽车股份有限公司 Vehicle-mounted microphone system and automobile

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102901950B (en) * 2012-09-20 2014-08-20 浙江工业大学 Method for recognizing three-dimensional coordinates of sound sources via planar arrays

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5867581A (en) * 1994-10-14 1999-02-02 Matsushita Electric Industrial Co., Ltd. Hearing aid
JP2010139476A (en) * 2008-12-15 2010-06-24 Nittobo Acoustic Engineering Co Ltd Calculation method of sound impedance, and system
JP2013015468A (en) * 2011-07-06 2013-01-24 Hitachi Engineering & Services Co Ltd Abnormal sound diagnostic apparatus and abnormal sound diagnostic method
JP2015059794A (en) * 2013-09-18 2015-03-30 株式会社小野測器 Sound source survey device
US11418871B2 (en) * 2018-05-04 2022-08-16 Sennheiser Electronic Gmbh & Co. Kg Microphone array
CN115150712A (en) * 2022-06-07 2022-10-04 中国第一汽车股份有限公司 Vehicle-mounted microphone system and automobile

Also Published As

Publication number Publication date
JP2760177B2 (en) 1998-05-28

Similar Documents

Publication Publication Date Title
US8577055B2 (en) Sound source signal filtering apparatus based on calculated distance between microphone and sound source
JP3863306B2 (en) Microphone array device
JPH08262121A (en) Position measuring instrument
JPH0572025A (en) Estimating method for position of sound source
JP2021034776A (en) Microphone location determination method, microphone array, and microphone system
JP5060465B2 (en) Sound collection device, sound collection method, sound collection program, recording medium
JP6659400B2 (en) Signal processing device, radar device, and method of setting radar device
JP4293986B2 (en) Method and system for representing a sound field
JP4787727B2 (en) Audio recording apparatus, method thereof, program thereof, and recording medium thereof
JP2011119898A (en) Device, method and program for acquiring sound
JP4928376B2 (en) Sound collection device, sound collection method, sound collection program using the method, and recording medium
JP2648110B2 (en) Signal detection method and device
US8606835B2 (en) Efficient kernel calculation for interpolation
JP2009005261A (en) Sound pickup apparatus, sound pickup method, sound pickup program using its method, and storage medium
CN106714192A (en) Network adjustment method and apparatus
US20070136377A1 (en) Optimum value search apparatus and method, recording medium, and computer program product
JP6253669B2 (en) Microphone device with improved directional characteristics
US10284988B2 (en) Method for analysing and decomposing stereo audio signals
JP2003143046A (en) Calibration device for receiving array antenna
JP2012037429A (en) Method for creating filter for beam-forming device and apparatus for creating filter for beam-forming device
JP2005265466A (en) Beam forming device based on minimax discipline, and its beam forming method
JP2746214B2 (en) Polygon mesh connection device
JP3582343B2 (en) Received signal processing device
JP5713933B2 (en) Sound source distance measuring device, acoustic direct ratio estimating device, noise removing device, method and program thereof
EP4354904A1 (en) Interpolation of finite impulse response filters for generating sound fields

Legal Events

Date Code Title Description
LAPS Cancellation because of no payment of annual fees