JP4972032B2 - Sound collecting device, sound collecting method, program thereof, and recording medium thereof - Google Patents

Description

  The present invention relates to, for example, a sound collection device that collects an acoustic signal emitted from a sound source, a sound collection method, a program thereof, and a recording medium thereof.

Conventionally, a microphone having a unidirectional characteristic is used in order to pick up an acoustic signal desired by a user with high sensitivity. For example, as described in Non-Patent Document 1, there are various types of microphones having unidirectional characteristics.
"Audio Visual Equipment Manual" Ritto Music Mook No. 44 April 26, 1996

  Conventional microphones with unidirectional characteristics have almost limited directivity direction, and the direction of the microphone when there are multiple sound sources (speakers) or when the sound source is biased to a certain region. If the user did not change the direction of directivity could not be changed.

  The sound collection device of the present invention includes an input unit, a switching processing unit, and an addition unit. The input unit is for selecting a left channel original signal that is the source of the left channel signal and a right channel original signal that is the source of the right channel signal from the collected sound signals collected by the four or more sound collecting units. Selection information is entered. The switching processing unit selects the left channel original signal and the right channel original signal using the selection information. The adding unit obtains a left channel signal by adding the respective left channel original signals, and obtains a right channel signal by adding the respective right channel original signals.

  The sound collection device of the present invention has a plurality of sound collection units (for example, microphones), and from these sound collection units, a left channel sound collection signal (hereinafter referred to as “left channel signal”). The directivity can be easily selected by manually or automatically selecting the sound collection unit that collects the sound and the sound collection unit that collects the sound as the right channel sound collection signal (hereinafter referred to as the “right channel signal”). The sound can be collected with high sensitivity.

  The best mode for carrying out the invention will be described below. In addition, the same number is attached | subjected to the process which performs the structure part which has the same function, and the same process, and duplication description is abbreviate | omitted.

FIG. 1 shows a functional configuration example of the sound collection device 100 according to the first embodiment. The sound collecting device of the present invention can easily change the directivity, and the sound collecting device 100 of the first embodiment manually changes the directivity. The sound collection device 100 includes a switching processing unit 4, an addition unit 6, and an input unit 8. The sound collection device 100 is connected to four or more sound collection units 2 _q (4 ≦ q ≦ Q, where q and Q are integers). The number of sound collection units 2 _q is preferably an even number, and the sound collection unit is preferably a unidirectional microphone, for example. In the following description, a case where there are four sound collection units (Q = 4) will be described. As shown in FIG. 1, the sound collecting portions 2 _q are arranged at equal angular intervals (90 degrees in the example of FIG. 1) along the circumference, and the front surfaces 2a _q (in FIG. clef 2 only _one of the front 2a ₁ shown), it is preferable to arranged so suited outwardly and in opposite directions to each other. The sound collecting apparatus of this embodiment is premised on collecting sound (stereo sound collecting) separately for the left channel and the right channel.

Here, “changing directivity” will be described. In the following description, the sound collection unit ₂ 1, _{2 2} in collected sound signal collected respectively (sound pickup unit ₂ 1, the output signal from the _{2 2)} P _{1, P} 2, (hereinafter, "left channel source signal "hereinafter.), respectively adding a left channel signal _{P L,} the sound pickup unit ₂ 3, _{2 4} collected sound signal collected respectively (sound pickup unit ₂ 3, the output signals from the _{2 4) P 3} , P _{4, (hereinafter,} referred to. "right channel original signal") to be processed in the state 1 to the right channel signal P _R by adding respectively. Also, the sound collection unit ₂ 1, _{2 4} collected signals _P 1, which is picked up by, _{P 4,} and a left-channel signal _{P L} by adding respectively, picked up by the sound pickup unit ₂ 2, _{2 3} yield sound signals P _2, P _3, and the processing of the state 2 to the right channel signal P _R by adding respectively. In the first embodiment, the user inputs selection information for determining whether to perform the state 1 process or the state 2 process from the input unit 8.

FIG. 2 schematically shows an example of directivity patterns A _{1 to} A ₄ (directivity patterns before change) inherent to each of the four sound collection units 2 ₁ to 2 ₄ with broken lines. In the following description, the “directivity pattern” is simply referred to as “directivity”. Furthermore, was synthesized sound pickup unit ₂ 1, _{2 2} of directional _A 1, _{A 2} in the case of the processing in the state 1 in Figure 2 (after the change) the directivity _{B L} and the sound pickup unit ₂ 3, _{2 4} the directional a _3, a ₄ were synthesized (after the change) the directivity B _R indicated by a one-dot chain line. 2, in the example of FIG. 3 to be described later, the sound collection unit ₂₁ to ₂₄ is a curve representing directivity A having originally assumed that the heart-shaped, the synthesized directivity B _L, the B _R The curve to represent is substantially elliptical. That is, in the state 1 processing, the sound collection units 2 ₁ and 2 ₂ are integrated to form a sound collection unit for the left channel signal (directivity is B _L ), and the sound collection units 2 ₃ and 2 ₄ are integrated. sound collection unit for the right channel signal (directivity B _R) is also said to form a.

Next, FIG. 3 shows the directivity _A 1 to A ₄ collecting sections 21 _{to 24} in the case of the processing condition 2 by broken lines, _(synthesized A 2, _{A 3)} directivity _B L after the change, B _R (synthesizes A ₁ and A ₄ ) is indicated by a one-dot chain line. In the case of the state 2 processing, as described above, the sound collection signals P ₁ and P ₄ collected by the sound collection units 2 ₁ and 2 ₄ are added to obtain the left channel signal P _L , and the sound collection unit 2 _{2, 2 3} collected signals _P 2, which is collected by, _{P 3,} respectively adds the right channel signal _{P R} and. That is, in the state 2 processing, the sound collection units 2 ₁ and 2 ₄ are integrated to form a sound collection unit for the left channel signal (directivity is B _L ), and the sound collection units 2 ₂ and 2 ₃ are integrated. sound collection unit for the right channel signal (directivity B _R) is also said to form the "end described in" changing the directionality "".

  In the sound collection device 100 according to the first embodiment, when the number of sound collection units is four, the user selects whether the state 1 (described in FIG. 2) or the state 2 (described in FIG. 3) is set (in the first embodiment, the user The directivity can be easily changed by selecting it by itself.

The processing flow will be described below with reference to FIG. FIG. 4 shows a processing flow.
First, selection information is input from the input unit 8 by the user (step S2). Here, the selection information refers to the left channel source signal and the right channel signal source, which are the sources of the left channel signal, from the collected sound signals collected by the Q sound collecting units 2 _q (Q = 4 in the above description). This is information for selecting the right channel original signal. 2 and 3, in the case of state 1, the left channel original signal corresponds to the sound pickup signals P ₁ and P ₂ , and the right channel original signal corresponds to the sound pickup signals P ₃ and P ₄ . . In the case of the state 2, the left channel original signal corresponds to the sound pickup signals P ₁ and P ₄ , and the right channel original signal corresponds to the sound pickup signals P ₂ and P ₃ .

The switching processing unit 4 uses the selection information input from the input unit 8 to select the left channel original signal and the right channel original signal (step S4). For example, when selection information regarding the processing in the state 1 is input from the input unit 8, a flag indicating that the signal is the left channel original signal is attached to the sound pickup signals P ₁ and P ₂ , and the sound pickup signal P ₃ is added. , it may lie down a flag indicating that the right channel based on signal to P _4.

Then, the adding unit 6 adds the left channel original signals (P ₁ + P _{2 in} the case of the state ₁ process, P ₁ + P _{4 in the} case of the process of state 2), and the left channel signal P _L. The right channel signal is obtained by adding the respective original right channel signals (P ₃ + P _{4 for} the state 1 process, P ₂ + P _{3 for} the state 2 process) (step S6). The left channel signal obtained, right channel signal P _R is output. That is, in the above example, the user can select the state 1 or the state 2 by inputting selection information to the input unit 8.

The effect of the sound collection device 100 will be described. In the following description, directional _B L, the area enclosed by the curve representing the _{B R} directional regions _C L, and _{C R.} Then, the directional region C _L, the acoustic signal from a sound source present in the C _R left channel signal, respectively, as a right channel signal, can be sensitively stereo sound pickup by picking up. As an explanation of the effect, consider two situations. First, as a first situation, there are two sound sources 20 ₁ and 20 ₂ (not shown). In this case, it picks up acoustic signals from example 20 ₁ as a left channel signal, picking up sound signals from 20 ₂ as a right channel signal, i.e., two sound sources 20 _1, 20 ₂ from the acoustic signal In the case where G ₁ and G ₂ are divided into left and right separate channels and stereo sound is collected, for example, when the user views the two sound sources 20 ₁ and 20 ₂ , an acoustic signal from the sound source 20 ₁ is input by inputting selection information. as the left channel signal, a right channel signal the sound signal from the sound source 20 _2, it can be picked up.

Next, the second situation is a situation in which the position of the sound source changes after a certain period of time. The change of the position of the sound source means, for example, that (i) the speaker moves or (ii) there are two speaker X and Y, and the speaker changes from X to Y. For example, as shown in FIG. 2, the sound source 20 (speaker) is present at the position α of the directional region C _L (i.e., the sound source in the vicinity of the sound pickup unit 2 _1, 2 ₂ are present ) When the processing in the state 1 is selected, the sound collection device 100 can collect the sound signal of the sound source 20 with high sensitivity as the left channel signal. However, when the sound source 20 moves to the position of β (in the vicinity of the sound collection units 2 ₁ and 2 ₄ in FIG. 2), if the processing of the state 1 is selected, there is a sound source outside the directivity region. The sound signal from the sound source cannot be collected with high sensitivity. Therefore, the user can easily change the directivity by switching from state 1 to state 2 (see FIG. 3) by inputting selection information, and as a result, the sound source 20 moves from α to β. However, the sound signal from the sound source 20 can be easily collected with high sensitivity.

  Conventionally, when the sound source 20 moves, the user himself has to change the direction of the microphone itself. However, with the sound collection device 100 according to the first embodiment, the user can easily change the directivity according to the selection information input by the user. Therefore, even if the sound source 20 moves, the sound collecting device 100 according to the first embodiment can easily collect the sound signal from the sound source 20 with high sensitivity.

  The sound collection device 100 described in the first embodiment changes the directivity by the user's manual operation (input of selection information), but the sound collection device 200 of the second embodiment estimates the position (direction) of the sound source. By doing so, the directivity is automatically changed according to the position of the sound source. FIG. 5 shows a functional configuration example of the sound collection device 200. The sound collection device 200 is different from the sound collection device 100 of the first embodiment in that the input unit 8 is replaced with the position estimation unit 30.

The sound collection signals P _{1 to} P ₄ collected by the four sound collection units 2 _{1 to} 2 ₄ are input to the switching processing unit 4 and the position estimation unit 30. The position estimation unit 30 compares the signal levels F _{1 to} F ₄ of the collected sound signals P _{1 to} P ₄ and determines that there is a sound source near the sound collecting unit that picks up the collected sound signal having the highest signal level. There is a technique. In the case of this method, for example, the position estimation unit 30 includes a signal level measurement unit 302, a signal level comparison unit 304, and a selection information generation unit 306.

The signal level measuring unit 302 measures the signal levels F _{1 to} F ₄ of the input sound pickup signals P _{1 to} P ₄ . The measured signal levels F _{1 to} F ₄ are input to the signal level comparison unit 304. The signal level comparison unit 304 compares the signal levels F _{1 to} F ₄ and selects the signal level having the highest level. Then, by determining that a sound source exists in the vicinity of the sound collection unit that has collected the sound collection signal having the highest signal level, position information indicating the position of the sound source is generated. The method for estimating the position of the sound source is not limited to the above, and a conventionally proposed position estimation method may be used. Further, the estimation of the positions of these sound sources can be applied even when there are two or more sound sources.

The selection information generation unit 306 generates selection information using the position information from the signal level comparison unit 304. As selection information described in Embodiment 1, the sound collection unit 21 _{to 24} from the sound collection signal P ₁ to P ₄ picked up by the left channel source signal and the right channel is the left channel signal P _L of the original is information for selecting the right channel original signal is the original signal P _R. There are various selection information generation criteria for the selection information generation means 306. Here, the selection information generation unit 306 will be described along the first and second situations.

First, a first situation (a situation where two sound sources 20 ₁ and 20 ₂ exist) will be described. For example, the signal level measuring means 302, the signal level comparing means 304, the sound source 20 ₁ is estimated to be present in the vicinity of the sound pickup unit _{2 1,} the sound source 20 ₂ is estimated to be present in the vicinity of the sound pickup unit _{2 2} Suppose. In this case, in order to effective stereo sound pickup, separately and collected signal P ₂ than collected signal P ₁ and the sound pickup unit 2 ₂ than sound pickup unit 2 ₁ to the right and left by channel sound As described above, the selection information generation unit 306 needs to generate selection information. For example, picked up the sound collection signal P ₁ as a left-channel signal P _L (i.e., the sound collection signal P ₁ is the left channel source signals), so as to pick up the sound pickup signal P ₂ as a right-channel signal P _R , (i.e., the sound collection signal P ₂ becomes right channel original signal) selection information generating means 306 may generate a selection information. Then, in the example of FIG. 5, the sound collection unit 2 ₁ is adjacent to the sound pickup unit 2 _4, the sound collection unit 2 ₂ is adjacent to the sound pickup unit 2 _3. Therefore, the sound collection unit ₂ 1, _{2 4} sound pickup signals _P 1, _{P 4} than the left channel source signal, sound pickup unit ₂ 2, _{2 3} from voice collecting signals _P 2, _{P 3} the right channel original signal The selection information generation unit 306 may generate selection information that is selected as The generated selection information is input to the switching processing unit 4.

The switching processing unit 4 selects the left channel original signal and the right channel original signal using the input selection information. In the above example, the switching processing unit 4, a sound collection signal _P 1, _{P 4} and left channel based signal, a sound pickup unit ₂ 2, ₂ collected signals _P 2, _{P 3} than ₃ as a right-channel original signal select. Then, the addition unit 6, each of the left channel original signal (in this example P _1, P ₄₎ determine the left channel signal P _L by adding the each of the right channel original signal (in this example P _2, P Request right channel signal P _R by adding the _3).

If, sound pickup unit ₂ 1, _{2 2} collected signals _P 1 than, _{P 2} becomes a left-channel signal are added is selected as the left channel source signal, sound pickup unit ₂ 3, ₂ collected signals than ₄ If P ₃ and P ₄ are sound collecting apparatuses that have only directivity (selected directivity is fixed) that are selected and added as the right channel original signal to become the right channel signal, depending on the position of the sound source, The stereo sound is not effectively collected, and the user needs to change the directivity by turning the sound collection unit or the like.

Next, the second situation (a situation where the position of the sound source changes after a certain period of time) will be described with reference to FIGS. As shown in FIGS. 2 and 3, even if the position of the sound source 20 moves from the position α to the position β, the signal level measuring means 302 and the signal level comparing means 304 in the position estimating unit 30 move the sound source 20. The position of the sound source can be estimated. For example, as shown in FIG. 2, the sound source 20 (speaker) is present at the position α of the directional region C _L (i.e., the sound source in the vicinity of the sound pickup unit 2 _1, 2 ₂ are present ), The acoustic signal G from the sound source 20 can be picked up with high sensitivity as the left channel signal by the selection information generated by the position estimation unit 30. Even when the sound source 20 has moved to the position β (in the vicinity of the sound collection units 4 ₁ , 4 ₄ in FIG. 2), the signal level measuring means 302 and the signal level comparing means 304 (β The position of the sound source 20 can be estimated. Therefore, the directivity is automatically changed by the selection information generated by the selection information generation means 306 as shown in FIG. Therefore, even if the sound source 20 moves, the sound collection device 200 can automatically collect the sound signal from the sound source 20 into either the left channel signal or the right channel signal, and the user cares about the directivity. Stereo sound can be collected without doing.

  Thus, with the sound collection device 200 of the second embodiment, the directivity can be automatically changed even when the sound source moves or there are a plurality of sound sources.

FIG. 6 shows a functional configuration example of the sound collection device 300 according to the third embodiment. The sound collection device 300 has a configuration in which a filter processing unit 40 is added to the sound collection device 100. Filter processing unit 40 serves to sharpen the formed directional B _L or B _R. 7, if the sound source 20 ₁ is in the position of gamma, a directional B _{R (indicated} by one-dot chain line), directional B _R which sharpened by the filter unit 40 'and a (indicated by two-dot chain line) Show. As shown in FIG. 7, the sound source 20 ₁ is sound pickup unit 2 ₁ around (hereinafter referred to as "position γ".) If there is in, the other around the sound pickup unit 2 ₄ forming the directivity B _R The formed directivity is reduced. In particular, for the sound pickup unit 2 ₄ backward directional locations (shown as E in FIG. 7) due to the sound pickup section 2 _4. In other words, hardly picks up acoustic signals from the rear of the sound pickup unit 2 _4, as a result, the sound signal from the sound source 20 allows sound pickup more stereo (stereo position憾) is clear.

Hereinafter, specific processing of the filter processing unit 40 will be described. In the following description, assuming that the two sound sources 20 ₁ and 20 ₂ are fixed, the acoustic signals from the sound sources 20 ₁ and 20 ₂ are represented as G ₁ and G ₂ , respectively. As shown in FIG. 6, the sound source 20 ₁ is in the vicinity of the sound pickup unit _{2 1,} the sound source 20 ₂ (not shown in FIG. 7) is to be near sound pickup unit _{2 3.} That is, the other collecting sections ₂ 2 sound pickup level of the acoustic signal _{G 1} is to be picked up by the sound pickup unit _{2 1,} 2 3, _{2 4,} in than sound pickup level of the acoustic signals _{G 1} to be picked up greater than it, the sound pickup unit _{2 3} sound pickup level of the audio signal _{G 2} is picked up by another sound pickup unit _{2 1,} 2 2, _{2 4,} in the collected by the sound collection level of the audio signal _{G 2} Also grows.

In the example of FIG. 6, the filter processing unit 40 is provided in the preceding stage of the switching processing unit 4, but may be provided in the subsequent stage of the switching processing unit 4 and the subsequent stage of the adding unit 6. The filter processing unit 40 of this embodiment includes a band dividing unit 402, an identifying unit 404, a weighting unit 406, and a combining unit 408. FIG. 8 shows a simplified process for the signal input by the filter processing unit 40. Each of the FIGS 8A~ Figure 8D, for sound collection signals _P 1 to P _4, band splitting means 402, the identifying means 404, describes processing performed by the weighting means 406. 8A to 8D show amplitude spectra of the collected sound signals P _{1 to} P ₄ converted into the frequency domain, respectively. The horizontal axis indicates the frequency, and the vertical axis indicates the power spectrum (level). Ω _m shown on the horizontal axis will be described later. 8E to 8H show waveforms of signals obtained by converting the frequency domain signals output from the weighting unit 406 into the time domain by the synthesizing unit 408 for the collected sound signals P _{1 to} P ₄ , respectively. The horizontal axis represents time t, and the vertical axis represents level.

The band dividing unit 402 converts each input signal into a frequency domain and divides it into frequency bands. In this example, “input signals” are the sound collection signals P _{1 to} P ₄ from the sound collection units 2 ₁ to 2 ₄ . Since the position where the filter processing unit 40 is provided may be a stage subsequent to the switching processing unit 4 or the like, the “input signal” varies depending on the position where the filter processing unit 40 is provided. Also, a known FFT (Fast Fourier Transform) or the like may be used as a method for transforming into the frequency domain. The number of frequency bands is M (M is an integer), and the frequency is expressed as ω _m (m = 1,..., M). Let X _q (ω _m ) be the frequency band signal that is the output signal from the band dividing means. However, as described above, q is an index of the sound collection unit (q = 1,..., Q, where Q = 4). Figure in the description of 8A~ Figure 8D, illustrating the acoustic signals _G 1, _{G 2} picked up for each band ω _m-1 ~ω _m. Here, as shown in FIG. 8A-D, an acoustic signal _G 1, _{G 2} explains why can be present separated (independent) within each band ω _m-1 ~ω _m. As a result of finely dividing the band of each collected sound signal, it is determined which sound source each band component is from, and all the determined components are output. That is, the sound source 20 _1, 20 if the frequency component of the acoustic signals G _1, G ₂ than ₂ do not overlap each other, for performing the process without missing a specific frequency band, while the sound source 20 ₁ maintaining high sound quality , it is possible to separate the acoustic signal from the 20 _2.

Then, the signal X _q (ω _m ) divided into frequency bands is input to the identification unit 404. The identification unit 404 identifies a high-level signal that is a signal having the highest level for each frequency band. In this example, at all frequencies, the power spectrum of the audio signal G ₁ is picked up by the sound pickup unit 2 ₁ (level) is the highest becomes (FIG. 8A), to identify the signal as a high-level signal. Similarly, the power spectrum of the audio signal G ₂ is picked up by the sound pickup unit 2 ₃ (level) is the highest becomes (FIG. 8C), identifying the signal as a high-level signal. Thus, for example, a flag may be attached to the identified high level signal.

The weighting means 406 weights so that the difference between the level of the high level signal and the level of other signals becomes large. Here, the “other signal” means another signal in the frequency band to which the high level signal belongs. For example, as a weighting method, a high level signal is multiplied by a weight value 1, and other signals are multiplied by a weight value α (0 <α <1) (see FIGS. 8A to 8D). Further, since the level difference between the high level signal and other signals only needs to be large, the weighting method is not limited. The weighting unit 406 obtains the weighting signal W _q (ω _m ) (q = 1 to 4) by performing weighting processing for each frequency band.

Then, the synthesizing unit 408 obtains a synthesized signal w _q (t) by converting the weighted signal W _q (ω _m ) in the frequency domain into the time domain. However, t shows time. For the time domain transform, inverse Fourier transform (IFFT) or the like may be used (see FIGS. 8E to 8H). By providing the filter processing unit 40 in this way, the directivity can be sharpened toward the side where the sound source 20 is located. In the above description, the case where there are two sound sources has been described. However, one or three or more sound sources can be applied.

The reason why the filter processing unit 40 performs processing in the frequency domain without performing processing in the time domain will be described. When the two sound sources ₂₀ 1, 20 ₂ are present, 20 levels of the audio signal _{G 1} than _1, consider the case is greater than the level of the audio signal _{G 2} than 20 _2. In this case, if the filter processing unit 40 performs processing in the time domain, the acoustic signal G ₂ having a small level is buried in the acoustic signal G ₁ having a large level, and the identifying unit 404 identifies the high-level signal. Because you will not be able to. Therefore, when there is one sound source, the identification means can identify the high level signal in each of the collected sound signals P _{1 to} P ₄ , and does not need to be converted to the frequency domain. That is, as shown in FIG. 9A, the band dividing unit 402 and the combining unit 408 can be omitted, and the identification unit 404 and the weighting unit 406 can be configured.

Further, a modification example of the filter processing unit 40 may be used as the filter processing unit 50 to perform processing using the phase difference between each of the collected sound signals P _{1 to} P ₄ . That is, using the phase difference, it can be seen that a sound source is present in the vicinity of the sound collecting portion where the acoustic signal has reached the earliest among the sound collecting portions 2 ₁ to 2 ₄ . FIG. 9B shows a functional configuration example of the filter processing unit using the phase difference. The filter processing unit 60 includes a phase difference measuring unit 410 and a weighting unit 406. First, the phase difference of each input signal is measured. And the sound collection part in which a sound source exists nearest from the said measurement result is recognized. Then, the weighting unit 406 performs weighting processing using the sound collection signal from the sound collection unit as a high level signal.

Consider a case where the sound source 20 has moved from the position of γ to the position of γ ′ shown in FIG. In this case, by performing the processing of the filter processing unit sequentially, it is possible to perform processing for sharpening directivity following the movement of the sound source. When the sound source 20 reaches the position γ ′, the directivity shown in FIG. 10 is formed.
Thus, by using the filter processing unit 40, the directivity can be sharpened corresponding to the position of the sound source. Even if the sound source 20 moves, the directivity can be sharpened following the movement of the sound source by sequentially performing the processing of the filter processing unit 40.

  FIG. 11 shows a functional configuration example of the sound collection device 400 according to the fourth embodiment. The sound collection device 400 has a configuration in which a filter processing unit 40 is provided before the switching processing unit 4 of the sound collection device 200. With such a configuration, the directivity is sharper than that of the sound collection device 200, and sound collection with a clear stereo feeling (sound localization feeling) is possible. The filter processing unit 40 may be provided before the switching processing unit 4, before the addition unit 6, or after the addition unit 6.

  In the above description, the number of sound collection units is four, but the number of sound collection units may be four or more. The directivity can be changed more finely as the number of sound collection units increases. Further, the number of the left channel original signal and the right channel original signal may not be the same. In this case, level adjustment may be performed after the addition processing by the adding unit 6 in order to adjust the level.

<Hardware configuration>
The present invention is not limited to the above-described embodiment. In addition, the various processes described above are not only executed in time series according to the description, but may be executed in parallel or individually according to the processing capability of the apparatus that executes the processes or as necessary. Needless to say, other modifications are possible without departing from the spirit of the present invention.
Further, when the above-described configuration is realized by a computer, the processing contents of the functions that the sound pickup apparatuses 100 to 400 should have are described by a program. The processing function is realized on the computer by executing the program on the computer.

  The program describing the processing contents can be recorded on a computer-readable recording medium. The computer-readable recording medium may be any medium such as a magnetic recording device, an optical disk, a magneto-optical recording medium, or a semiconductor memory. Specifically, for example, the magnetic recording device may be a hard disk device or a flexible Discs, magnetic tapes, etc. as optical disks, DVD (Digital Versatile Disc), DVD-RAM (Random Access Memory), CD-ROM (Compact Disc Read Only Memory), CD-R (Recordable) / RW (ReWritable), etc. As the magneto-optical recording medium, MO (Magneto-Optical disc) or the like can be used, and as the semiconductor memory, EEP-ROM (Electronically Erasable and Programmable-Read Only Memory) or the like can be used.

  The program is distributed by selling, transferring, or lending a portable recording medium such as a DVD or CD-ROM in which the program is recorded. Furthermore, the program may be distributed by storing the program in a storage device of the server computer and transferring the program from the server computer to another computer via a network.

  A computer that executes such a program first stores, for example, a program recorded on a portable recording medium or a program transferred from a server computer in its own storage device. When executing the process, the computer reads a program stored in its own recording medium and executes a process according to the read program. As another execution form of the program, the computer may directly read the program from a portable recording medium and execute processing according to the program, and the program is transferred from the server computer to the computer. Each time, the processing according to the received program may be executed sequentially. Also, the program is not transferred from the server computer to the computer, and the above-described processing is executed by a so-called ASP (Application Service Provider) type service that realizes the processing function only by the execution instruction and result acquisition. It is good. Note that the program in this embodiment includes information that is used for processing by an electronic computer and that conforms to the program (data that is not a direct command to the computer but has a property that defines the processing of the computer).

In this embodiment, the present apparatus is configured by executing a predetermined program on a computer. However, at least a part of these processing contents may be realized by hardware.
Further, the word adding device described in the present embodiment includes a CPU (Central Processing Unit), an input unit, an output unit, an auxiliary storage device, a RAM (Random Access Memory), a ROM (Read Only Memory), and a bus. (Neither shown).

  The CPU executes various arithmetic processes according to the read various programs. The auxiliary storage device is, for example, a hard disk, an MO (Magneto-Optical disc), a semiconductor memory, or the like, and the RAM is an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), or the like. The bus connects the CPU, the input unit, the output unit, the auxiliary storage device, the RAM, and the ROM so that they can communicate with each other.

<Cooperation between hardware and software>
The word adding device of this embodiment is constructed by reading a predetermined program into the hardware as described above and executing it by the CPU. The functional configuration of each device constructed in this way will be described below.
The input unit 8 is a communication device such as a LAN card or a modem that is driven under the control of a CPU loaded with a predetermined program. The switching processing unit, the adding unit, the position estimating unit, and the filter processing unit are arithmetic units that are constructed by reading a predetermined program into the CPU and executing it.

2 is a functional configuration example of the sound collection device according to the first embodiment. The figure which shows the directivity formed in the case of the process of the state 1. FIG. The figure which shows the directivity formed in the case of the process of the state 2. FIG. The figure which showed the processing flow of the sound collection device of a present Example. 6 is a functional configuration example of a sound collection device according to Embodiment 2. FIG. 9 is a functional configuration example of a sound collection device according to a third embodiment. The figure which shows the directivity sharpened by the filter process part. The figure which showed the process of the filter process part. 9A is a diagram illustrating a functional configuration example of the filter processing unit 50, and FIG. 9B is a diagram illustrating a functional configuration example of the filter processing unit 60. The figure which shows the directivity when a sound source moves to the position of (gamma) '. FIG. 6 is a diagram illustrating a functional configuration example of a sound collection device according to a fourth embodiment.

Claims (10)

Selection information for selecting the left channel original signal that is the source of the left channel signal and the right channel original signal that is the source of the right channel signal is input from the collected sound signals collected by the four or more sound collection units. An input unit,
A switching processing unit that selects a left channel original signal and a right channel original signal using the selection information;
A sound collecting apparatus comprising: an adder that obtains the left channel signal by adding each left channel original signal and obtains the right channel signal by adding each right channel original signal. By estimating the position of the sound source, the position information of the position of the sound source is obtained, and the position information is used as the source of the left channel signal from the collected sound signals collected by four or more sound collecting units. A position estimator that generates selection information for selecting a right channel original signal that is the source of the left channel original signal and the right channel signal;
A switching processing unit that selects a left channel original signal and a right channel original signal using the selection information;
A sound collecting apparatus comprising: an adder that obtains the left channel signal by adding each left channel original signal and obtains the right channel signal by adding each right channel original signal. The sound collecting device according to claim 1 or 2,
The sound collecting device further includes a filter processing unit that sharpens directivity using the input signal. The sound collection device according to claim 3,
The filter processing unit
Band division means for converting each input signal into a frequency domain and dividing it into frequency bands;
Identifying means for identifying a high-level signal that is a signal having the highest level for each frequency band;
For each frequency band, a weighting unit that obtains a weighting signal by weighting so that a difference between the level of the high-level signal and the level of another signal becomes large;
Combining means for converting the weighted signal into a time domain;
A sound collecting device comprising: Selection information for selecting the left channel original signal that is the source of the left channel signal and the right channel original signal that is the source of the right channel signal is input from the collected sound signals collected by the four or more sound collection units. Input process
Using the selection information to select a left channel source signal and a right channel source signal; and
A sound collection method comprising: an adding step of obtaining the left channel signal by adding each left channel original signal and obtaining the right channel signal by adding each right channel original signal. By estimating the position of the sound source, the position information of the position of the sound source is obtained, and the position information is used as the source of the left channel signal from the collected sound signals collected by four or more sound collecting units. A selection information generation process for generating selection information for selecting a right channel original signal that is an origin of the left channel original signal and the right channel signal;
A selection process for selecting a left channel original signal and a right channel original signal using the selection information;
A sound collection method comprising: an addition step of obtaining the left channel signal by adding each left channel original signal and obtaining the right channel signal by adding each right channel original signal. The sound collection method according to claim 5 or 6,
The sound collection method further includes a filtering process for sharpening directivity using the input signal. The sound collection method according to claim 7,
The filtering process includes:
A band division step for converting each input signal into a frequency domain and dividing it into frequency bands;
An identification step for identifying a high-level signal that is a signal having the highest level for each frequency band; and
For each frequency band, a weighting step for obtaining a weighting signal by weighting so that a difference between the level of the high-level signal and the level of another signal becomes large;
Combining the weighted signal into a time domain;
A sound collection method comprising:   A program for operating a computer as the sound collecting device according to claim 1.   A computer-readable recording medium on which the program according to claim 9 is recorded.

Images