close

Вход

Забыли?

вход по аккаунту

?

JP2009100072

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2009100072
An object of the present invention is to make it possible to determine abnormal operation of a
plurality of microphones constituting a microphone array even during voice processing.
SOLUTION: A silence judgment unit 1241 judges silence, and a microphone array abnormal
operation judgment unit 124 does not carry out microphone array abnormal operation judgment
processing when the silent judgment unit 1241 judges that there is no sound. If not silent, the
microphone array abnormal operation determination unit 124 determines the microphone array
abnormal operation based on the signal level difference between the signal levels of the digital
audio signals corresponding to the microphones 1211 to 1213 and the correlation coefficient.
The control unit 17 changes the microphones 1211 to 1213 used for audio processing so as not
to use the microphone for abnormal operation for audio processing when notified of the
microphone array abnormal operation etc. from the microphone array abnormal operation
determination unit 124 I do. [Selected figure] Figure 1
Voice processing apparatus, voice processing method and program for voice processing
apparatus
[0001]
The present invention relates to an audio processing device, an audio processing method of the
audio processing device, and a program.
[0002]
Some voice processing devices perform signal processing so as not to receive unnecessary sound
as much as possible by using a microphone array configured by a plurality of microphones, and
such voice processing devices are widely used in television conference systems and the like. (See,
04-05-2019
1
for example, Non-Patent Document 1).
[0003]
In such a voice processing apparatus, when a failure or the like occurs in any of the microphones,
normal voice processing can not be performed.
[0004]
For this reason, when the input signal does not have a power higher than a predetermined level,
the abnormal operation determination of each microphone is performed by determining that the
microphone is a failure, and signal processing is performed using only the microphones
diagnosed as normal. There is an audio processing device (see, for example, Patent Document 1).
Japanese Patent Application Laid-Open No. 2002-159086 JP Togao Oga, Yoshio Yamazaki,
Toyoda Kanada
[0005]
However, in the conventional voice processing apparatus, since an abnormal operation is
determined when the signal after the input does not have the power of the predetermined level
or more, it is necessary to externally input the signal having the power of the predetermined
level or more from outside. For example, abnormal operation determination can not be
performed.
[0006]
For this reason, in the conventional voice processing apparatus, failure diagnosis can not be
performed during voice processing in which an arbitrary voice signal is input, and a voice signal
for abnormal operation determination is externally input during a period in which voice
processing is not performed. It is necessary to determine abnormal operation.
[0007]
The present invention has been made in view of such conventional problems, and an audio
processing apparatus capable of determining abnormal operation of a plurality of microphones
constituting a microphone array even during audio processing. Intended to be provided.
04-05-2019
2
[0008]
Another object of the present invention is to provide an audio processing apparatus capable of
continuing normal audio signal processing even during audio processing, and an audio
processing method and program for the audio processing apparatus.
[0009]
In order to achieve this object, a voice processing apparatus according to a first aspect of the
present invention uses a microphone array configured of a plurality of microphones for inputting
voice and a voice input by the plurality of microphones of the microphone array. Voice acquiring
means for acquiring each audio signal, comparing feature amounts of the acquired audio signals,
and determining an operation abnormality of each microphone of the microphone array based on
a comparison result; And a voice processing unit configured to determine a microphone to be
used for voice processing based on the determination result of the means and to perform voice
processing using the determined microphone.
[0010]
The voice acquisition means includes a silence determination means for comparing the signal
level of each voice signal obtained with the silence determination value set in advance and
determining silence based on the comparison result, the silence determination means comprising
silence When determined, the operation abnormality of each microphone of the microphone
array may not be determined, and the operation abnormality of each microphone may be
determined when it is determined that the silence determination unit is not silence.
[0011]
When the silence determination means compares the signal level of each voice signal acquired by
the voice acquisition means with the preset silence determination value, and the signal levels of
all the voice signals become less than the silence determination value Alternatively, time may be
measured in which the signal level of all the audio signals is less than the silence determination
value, and may be determined as silence when the measurement time exceeds a predetermined
time set in advance.
[0012]
The sound acquisition unit may compare signal levels of the sound signals as feature amounts of
the acquired sound signals, and determine an operation abnormality of each microphone of the
microphone array based on a comparison result.
04-05-2019
3
[0013]
The voice acquisition unit acquires, as the feature quantity, the correlation coefficient of the
amplitude characteristic of each acquired audio signal, compares the acquired correlation
coefficient, and based on the comparison result, the operation abnormality of each microphone
of the microphone array May be determined.
[0014]
A confirmation sound output means is provided for outputting a confirmation sound at which
voice feature amount is set in advance at the start of speech processing, and the plurality of
microphones of the microphone array input the confirmation sound outputted by the
confirmation sound output means. The voice acquiring unit acquires respective audio signals
based on confirmation sounds respectively input by the plurality of microphones of the
microphone array, compares feature amounts of the acquired audio signals, and based on a
comparison result. The operation abnormality of each microphone of the microphone array may
be determined at the start of the audio processing.
[0015]
A warning sound output unit may be provided that outputs a warning sound when the sound
acquisition unit determines that the microphone operation of the microphone array is abnormal.
[0016]
A voice processing method of a voice processing device according to a second aspect of the
present invention is a voice processing method of a voice processing device including a
microphone array configured of a plurality of microphones for inputting voice, the voice
processing method of the microphone array The step of acquiring each audio signal based on the
voice input by each of the plurality of microphones and the feature amount of each acquired
audio signal are compared, and based on the comparison result, the operation abnormality of
each microphone of the microphone array is And determining the microphone to be used for
voice processing based on the determination result of the operation abnormality of the
microphone, and performing voice processing using the determined microphone. Do.
[0017]
A program according to a third aspect of the present invention is a program for acquiring a voice
signal of each of a plurality of microphones constituting a microphone array based on voices
input to the computer, and acquiring feature amounts of the obtained voice signals. A procedure
of comparing and determining an operation abnormality of each microphone of the microphone
04-05-2019
4
array based on a comparison result, a procedure of determining a microphone to be used for
voice processing based on a determination result of determining the operation abnormality of the
microphone, A procedure for performing audio processing using a microphone is performed.
[0018]
According to the present invention, it is possible to determine the abnormal operation of a
plurality of microphones constituting the microphone array even during audio processing.
In addition, even during audio processing, normal audio signal processing can be continued.
[0019]
Hereinafter, a speech processing apparatus according to an embodiment of the present invention
will be described with reference to the drawings.
The configuration of the speech processing apparatus according to the present embodiment is
shown in FIG.
The voice processing device 1 according to the present embodiment includes an information
storage unit 11, a voice acquisition unit 12, a voice output unit 13, an operation unit 14, a
display unit 15, a communication unit 16, and a control unit 17. Prepare.
[0020]
The information storage unit 11 stores a program, user information, information received by the
communication unit 16, voice data, a microphone variation correction value 111, a standard
value, and the like.
[0021]
The microphone variation correction value 111 is a correction value for correcting the
characteristics of a plurality of microphones.
04-05-2019
5
[0022]
The microphone variation correction value 111 is written to the information storage unit 11 at
the time of product shipment or the like based on the microphone characteristic data measured
in advance.
[0023]
The standard value is various threshold values used for silence determination, microphone array
abnormal operation determination, and specification of a microphone of operation abnormality.
[0024]
The voice acquisition unit 12 acquires a plurality of voice signals, and includes a microphone
array 121, an A / D converter, a microphone deviation correction unit 123, and a microphone
array abnormal operation determination unit 124.
[0025]
Further, the voice acquisition unit 12 obtains a plurality of voice signals, compares the feature
amounts of the obtained voice signals, and determines the microphone array abnormal operation
based on the comparison result.
[0026]
The voice processing device 1 is provided with such a voice acquisition unit 12 so that abnormal
operation determination of a plurality of microphones constituting the microphone array can be
performed even during voice processing.
[0027]
The microphone array 121 is configured by three microphones 1211, 1212, 1213 as a plurality
of microphones.
The microphones 121, 121, 1213 respectively receive voice and output analog voice signals
Sma1, Sma2, Sma3.
04-05-2019
6
[0028]
The A / D converter performs analog-digital conversion on the analog audio signals Sma1, Sma2,
and Sma3 output from the microphones 1211, 1212, and 1213, respectively, and outputs digital
audio signals Smd1, Smd2, and Smd3.
[0029]
The microphone dispersion correction unit 123 corrects the characteristic dispersion of the
microphones 12 11 12 12 12 13 using the microphone dispersion correction value 111 stored
in the information storage unit 11.
[0030]
The microphone variation correction unit 123 corrects the amplitude characteristics, phase
characteristics, and the like of the microphones 1211, 1212, and 1213 as characteristic
variations.
[0031]
The microphone array abnormal operation judging unit 124 judges the microphone array
abnormal operation by the failure of the microphones 121, 121, 1213, the microphone blocking
and the like.
[0032]
The voice processing device 1 is provided with input holes for inputting voices to the
microphones 121, 121, 1213, respectively. The microphone closing means that the input holes
are closed.
[0033]
If any one of the microphones 121, 121, 1213 has a failure or a microphone obstruction, the
voice processing device 1 can not perform normal voice processing using all the microphones
1211, 1212, 1213.
[0034]
The microphone array abnormal operation determination unit 124 compares the feature
amounts of the digital audio signals Smd1, Smd2, and Smd3 and determines such an operation
state as the microphone array abnormal operation based on the comparison result.
04-05-2019
7
[0035]
Therefore, the microphone array abnormal operation determination unit 124 includes a silence
determination unit 1241, a volume comparison unit 1242, and a correlation coefficient
calculation unit 1243.
[0036]
The silence determination unit 1241 performs silence determination.
Silence means a state in which no sound is input to all the microphones 1211, 1212, 1213.
[0037]
The silence determination is performed so that all microphones 1211 to 1213 are not
erroneously determined as abnormal operation in this silent period even when a voice signal
having a silent period is selected as an external input signal for abnormal operation
determination. In order to
[0038]
The silence determination unit 1241 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 of
the digital audio signals Smd1, Smd2, and Smd3 with the prescribed level Lmd_th set in advance
to determine silence.
[0039]
The information storage unit 11 stores the predetermined level Lmd_th in advance, and the
silence determination unit 1241 reads the predetermined level Lmd_th from the information
storage unit 11 and compares it with the signal levels Lmd_1, Lmd_2, and Lmd_3.
[0040]
The silence determination unit 1241 has a timer for measuring time (not shown), and measures
the time from when the signal level Lmd_1, Lmd_2, and Lmd_3 is less than the defined level
Lmd_th, when it is less than the defined level Lmd_th.
04-05-2019
8
Let T be this measurement time.
[0041]
The information storage unit 11 stores in advance a specified time Tth to be compared with the
measurement time T.
The silence determination unit 1241 reads the specified time Tth from the information storage
unit 11 and compares the measured time T with the specified time Tth.
[0042]
The silence determination unit 1241 determines that no sound occurs when the measurement
time T exceeds the specified time Tth, and the microphone array abnormal operation
determination unit 124 determines that the microphone array abnormal operation occurs when
the silence determination unit 1241 determines that no sound occurs. Do not do.
[0043]
The volume comparing unit 1242 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 as
feature quantities of the digital audio signals Smd1, Smd2, and Smd3 in order to determine the
microphone array abnormal operation.
[0044]
For example, when the signal level difference ΔL_12 between the signal levels Lmd_1 and
Lmd_2 is very large compared to the signal level difference ΔL_23 between the signal levels
Lmd_2 and Lmd_3 and the signal level difference ΔL_31 between the signal levels Lmd_3 and
Lmd_1, the microphone 1211 is There may be a failure or there may have been a microphone
obstruction to the microphone 1211.
[0045]
Therefore, the volume comparing unit 1242 compares the signal levels Lmd_1, Lmd_2, and
Lmd_3 to obtain the signal level differences ΔL_12, ΔL_23, and ΔL_31.
04-05-2019
9
[0046]
The correlation coefficient calculation unit 1243 describes the cross correlation coefficient
between the digital audio signals Smd1 and Smd2 (hereinafter simply referred to as “correlation
coefficient”) as a feature of the digital audio signals Smd1, Smd2 and Smd3.
) R12, a correlation coefficient r23 between the digital audio signals Smd2 and Smd3, and a
correlation coefficient r31 between the digital audio signals Smd3 and Smd1.
[0047]
For example, when the correlation coefficient r12 between the signal levels Lmd_1 and Lmd_2
and the correlation coefficient r31 between the signal levels Lmd_3 and Lmd_1 are very small in
comparison with the correlation coefficient r23 between the signal levels Lmd_2 and Lmd_3, the
microphone 1211 There may be a failure or there may have been a microphone obstruction to
the microphone 1211.
[0048]
The correlation coefficient calculation unit 1243 calculates the correlation coefficients r11, r23,
r31 to determine such an abnormal operation of the microphone array.
[0049]
The correlation coefficient calculation unit 1243 accumulates the signal levels Lmd_1, Lmd_2,
and Lmd_3 of the digital audio signals Smd1, Smd2, and Smd3 for a prescribed period set in
advance.
[0050]
The correlation coefficient calculation unit 1243 performs Fourier transform on the accumulated
signal levels Lmd_1, Lmd_2, and Lmd_3, and calculates correlation coefficients r11, r23, and r31
for each obtained amplitude characteristic.
[0051]
When the volume comparison unit 1242 and the correlation coefficient calculation unit 1243
perform such calculations, the microphone array abnormal operation determination unit 124
04-05-2019
10
determines based on the calculation result of the volume comparison unit 1242 and the
calculation result of the correlation coefficient calculation unit 1243. The microphone array
abnormal operation is determined.
[0052]
When using the calculation result of the volume comparison unit 1242, the microphone array
abnormal operation determination unit 124 compares each of the signal level differences
ΔL_12, ΔL_23, ΔL_31 acquired by the volume comparison unit 1242 with the preset standard
value ΔL_th. .
[0053]
The standard value ΔL_th is a threshold value preset for the microphone array abnormal
operation determination, and the information storage unit 11 stores the standard value ΔL_th.
The microphone array abnormal operation determination unit 124 reads the standard value
ΔL_th from the information storage unit 11 and compares it with the signal level differences
ΔL_12, ΔL_23, and ΔL_31, respectively.
[0054]
When all the signal level differences ΔL_12, ΔL_23, ΔL_31 are less than the standard value
ΔL_th as a result of comparison, the microphone array abnormal operation determination unit
124 determines that the microphone array 121 is operating normally.
[0055]
On the other hand, if there is a signal level difference ΔL_12, ΔL_23, ΔL_31 which is equal to
or larger than the standard value ΔL_th, the microphone array abnormal operation judging unit
124 judges that the microphone array abnormal operation.
[0056]
When it is determined that the microphone array abnormal operation is performed, the
microphone array abnormal operation determination unit 124 compares the signal levels Lmd_1,
Lmd_2, and Lmd_3 to determine a microphone with an operation abnormality.
04-05-2019
11
[0057]
Since the voices input to the microphones 1211 to 1213 fluctuate, the microphone array
abnormal operation determination unit 124 determines the microphone with operation
abnormality based on the highest signal level among the signal levels Lmd_1, Lmd_2, and
Lmd_3.
[0058]
That is, assuming that the highest signal level is Lmd_x, the other signal levels are Lmd_y, and
the signal level difference between the signal levels Lmd_x and Lmd_y is ΔL_xy, the microphone
array abnormal operation determination unit 124 determines the signal level difference ΔL_xy
and the specified value ΔL_th. Compare.
[0059]
When the signal level difference ΔL_xy is equal to or larger than the specified value ΔL_th, the
microphone array abnormal operation determination unit 124 determines that the microphone
corresponding to the signal level Lmd_y is abnormal in operation.
[0060]
When the calculation result of the correlation coefficient calculation unit 1243 is used, the
microphone array abnormal operation determination unit 124 sets each of the correlation
coefficients r11, r23, r31 calculated by the correlation coefficient calculation unit 1243 to a
preset standard value. Compare with r_th.
[0061]
The standard value r_th is a threshold preset for the microphone array abnormal operation
determination, and the information storage unit 11 stores the standard value r_th.
The microphone array abnormal operation determination unit 124 reads the standard value r_th
from the information storage unit 11 and compares it with the correlation coefficients r11, r23,
and r31, respectively.
[0062]
04-05-2019
12
The microphone array abnormal operation determination unit 124 determines that the
microphones 1211 to 1213 operate normally if all of the correlation coefficients r11, r23, and
r31 are equal to or greater than the standard value r_th.
[0063]
On the other hand, if the correlation coefficient r11 is less than the standard value r_th, the
microphone array abnormal operation determination unit 124 determines that the correlation
coefficient r11 is low.
Similarly, if the correlation coefficients r23 and r31 are less than the standard value r_th,
respectively, the microphone array abnormal operation determination unit 124 determines that
the correlation coefficients r23 and r31 are low.
[0064]
In the case of the correlation coefficient, for example, when the microphones 1212 and 1213 are
normal and the microphone 1211 is broken or there is a microphone obstruction, the correlation
coefficients r12 and r31 become low.
[0065]
In this case, since the correlation coefficients r12 and r31 are correlation coefficients related to
the microphone 1211, the microphone array abnormal operation determination unit 124
determines that the microphone array abnormal operation is performed, and determines that the
microphone 1211 is an operation abnormal microphone.
[0066]
In this embodiment, the microphone array abnormal operation judging unit 124 judges the
microphone array abnormal operation using both the calculation result of the volume comparing
unit 1242 and the calculation result of the correlation coefficient calculating unit 1243. Do.
[0067]
When the microphone array abnormal operation determination unit 124 performs the
04-05-2019
13
microphone array abnormal operation determination process, the microphone array abnormal
operation determination unit 124 notifies the control unit 17 of the determination result.
When it is determined that the microphone array abnormal operation is performed, the control
unit 17 is notified of the microphone which has become an operation abnormality together with
the determination result of the microphone array abnormal operation.
[0068]
The voice output unit 13 outputs a voice for notifying the user of the voice processing device 1
of the state of the voice processing device 1.
The audio output unit 13 outputs a confirmation sound at the start of audio processing.
[0069]
Further, the voice output unit 13 outputs a warning sound to notify the user of the voice
processing apparatus 1 that the microphone array abnormal operation factor is present at the
time of determination of the microphone array abnormal operation.
[0070]
The operation unit 14 receives an input of the user of the voice processing device 1.
The operation unit 14 includes a plurality of operation keys (not shown) operated by the user,
and receives a start instruction, an end instruction and the like of voice processing inputted by
the user operating the plurality of operation keys.
When receiving the user's input, the operation unit 14 supplies the input information to the
control unit 17.
[0071]
04-05-2019
14
The display unit 15 displays information.
The display unit 15 displays the information received by the communication unit 16, the
information stored in the information storage unit 11, and the information input by the user of
the voice processing device 1 to the operation unit 14.
[0072]
The communication unit 16 communicates with other devices via a network.
When the audio data is supplied from the control unit 17, the communication unit 16 transmits
the supplied audio data.
[0073]
The control unit 17 includes a central processing unit (CPU), a read only memory (ROM), a
random access memory (RAM), and the like to control each unit of the voice processing
apparatus 1.
[0074]
The control unit 17 receives input information of the start of the audio processing from the
operation unit 14 and starts the audio processing.
Further, the control unit 17 receives the input information of the end command from the
operation unit 14 and ends the voice processing.
[0075]
Also, the control unit 17 includes an audio processing unit 171.
04-05-2019
15
The control unit 17 determines (sets) a microphone to be used for audio processing based on the
determination result of the microphone array abnormal operation determination unit 124 of the
audio acquisition unit 12 and performs audio processing using the determined microphone. The
voice processing unit 171 is controlled.
[0076]
The voice processing unit 171 performs voice processing and voice processing on the digital
voice signals Smd1, Smd2, and Smd3 acquired by the voice acquisition unit 12, and includes an
echo canceller processing unit 1711, a noise removal processing unit 1712, and a sound source.
A localization processing unit 1713, a sound source separation processing unit 1714, and a voice
recognition processing unit 1715 are provided.
[0077]
The echo canceller processing unit 1711 performs echo canceller processing by referring to any
one of the digital audio signals Smd1, Smd2 and Smd3 as a reference signal of the echo canceller.
[0078]
The echo canceller processing unit 1711 suppresses the output voice component of the voice
output unit 13 from the digital voice signals Smd1, Smd2, and Smd3 input from the voice
acquisition unit 12 by performing such echo canceller processing.
[0079]
The noise removal processing unit 1712 performs standing noise removal and filtering
processing of a specific frequency estimated to be noise.
[0080]
The sound source localization processing unit 1713 performs sound source localization
processing.
[0081]
The sound source separation processing unit 1714 performs sound source separation
processing.
04-05-2019
16
[0082]
The voice recognition processing unit 1715 performs voice recognition.
[0083]
When the control unit 17 is supplied with input information indicating that communication is to
be performed from the operation unit 14, the control unit 17 supplies to the communication unit
16 the digital sound signal subjected to the sound processing process by the sound processing
unit 171 and stores the signal. When the input information is supplied, the digital audio signal is
stored in the information storage unit 11.
[0084]
Next, the operation of the speech processing device 1 according to the present embodiment will
be described.
When the control unit 17 receives input information of an instruction to start voice processing
from the operation unit 14, the control unit 17 reads out program data of voice processing from
the information storage unit 11, and performs voice processing according to the flowcharts
shown in FIGS. Run.
[0085]
The control unit 17 causes the voice output unit 13 to output a confirmation sound for notifying
the user of the voice processing apparatus 1 of the start of processing (step S101).
[0086]
The control unit 17 instructs the voice acquisition unit 12 to execute the voice acquisition
process (step S102).
The voice acquisition unit 12 executes this voice acquisition process according to the flowchart
shown in FIG.
04-05-2019
17
[0087]
First, the microphones 1211 to 1213 of the voice acquisition unit 12 acquire the analog voice
signals Sma1, Sma2, and Sma3 (step S121).
[0088]
The A / D converter 122 converts the analog audio signals Sma1, Sma2 and Sma3 obtained by
the microphones 1211 to 1213 into digital audio signals Smd1, Smd2 and Smd3 (step S122).
[0089]
The microphone variation correction unit 123 acquires the microphone variation correction
value 111 from the information storage unit 11, and performs the microphone variation
correction based on the acquired microphone variation correction value 111 (step S123).
Then, the voice acquisition unit 12 ends the voice acquisition process.
[0090]
When the voice acquisition unit 12 executes the voice acquisition process, the control unit 17
instructs the microphone array abnormal operation determination unit 124 of the voice
acquisition unit 12 to execute the microphone array abnormal operation determination process
(step S103).
[0091]
The microphone array abnormal operation judging unit 124 executes the microphone array
abnormal operation judging process according to the flowchart shown in FIG.
[0092]
The silence determination unit 1241 of the microphone array abnormal operation determination
unit 124 reads out the specified level Lmd_th from the information storage unit 11 and
compares it with the digital audio signals Smd1, Smd2 and Smd3 to perform silence
determination (steps S201 to S203).
04-05-2019
18
[0093]
That is, the silence determination unit 1241 compares the signal levels Lmd_1, Lmd_2, Lmd_3
with the defined level Lmd_th, and all the signal levels Lmd_1, Lmd_2, Lmd_3 of the digital audio
signals Smd1, Smd2, Smd3 are less than the defined level Lmd_th. It is determined whether or
not (step S201).
[0094]
When it is determined that the signal levels Lmd_1, Lmd_2, and Lmd_3 of the digital audio
signals Smd1, Smd2, and Smd3 are all less than the defined level Lmd_th (Step S201; Yes), the
silence determination unit 1241 measures the time from this time ( Step S202).
[0095]
The silence determination unit 1241 determines whether the measurement time T has exceeded
a predetermined time Tth set in advance (step S203).
[0096]
If it is determined that the measurement time T is less than or equal to the preset specified time
Tth (step S203; No), the silence determination unit 1241 determines again whether all the signal
levels Lmd_1, Lmd_2, and Lmd_3 are less than the specified level Lmd_th. It is determined (step
S201).
[0097]
When the silence determination unit 1241 determines that the signal levels Lmd_1, Lmd_2, and
Lmd_3 are all less than the prescribed level Lmd_th, and the measurement time T exceeds the
prescribed time Tth (step S203; Yes), the microphone array abnormal operation determiner 124
The microphone array abnormal operation determination process is ended.
[0098]
On the other hand, when at least one of the signal levels Lmd_1, Lmd_2, Lmd_3 is equal to or
higher than the prescribed level Lmd_th, the silence determination unit 1241 determines that the
signal is not silent (Step S201; No).
[0099]
When the microphone array abnormal operation determination unit 124 determines that the
04-05-2019
19
silence determination unit 1241 is not silence, causes the volume comparison unit 1242 to
compare the signal levels Lmd_1, Lmd_2, and Lmd_3 of the digital audio signals Smd1, Smd2,
and Smd3 (step S205).
[0100]
The volume comparing unit 1242 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 to
obtain signal level differences ΔL_12, ΔL_23, and ΔL_31 between the digital audio signals
Smd1 and Smd2.
[0101]
The microphone array abnormal operation determination unit 124 causes the correlation
coefficient calculation unit 1243 to calculate the correlation coefficients r12, r23, r31 among the
digital audio signals Smd1, Smd2, Smd3 (step S205).
[0102]
The microphone array abnormal operation determination unit 124 is a microphone array based
on the signal level differences ΔL_12, ΔL_23, ΔL_31 acquired by the volume comparison unit
1242, and the correlation coefficients r12, r23, r31 calculated by the correlation coefficient
calculation unit 1243. It is determined whether or not it is an abnormal operation (step S206).
[0103]
When it is determined that the microphones 1211 to 1213 are operating normally (step S206;
No), the microphone array abnormal operation determination unit 124 notifies the control unit
17 of no microphone array abnormal operation (step S207), The microphone array abnormal
operation determination process is ended.
[0104]
On the other hand, when the microphone array abnormal operation is determined (step S206;
Yes), the microphone array abnormal operation determination unit 124 compares the signal level
differences ΔL_12, ΔL_23, ΔL_31, or the correlation coefficients r12, r23, r31, The
microphone of the abnormal operation is determined (step S208).
[0105]
The microphone array abnormal operation determination unit 124 notifies the control unit 17 of
the determination result of the microphone array abnormal operation and the microphones of
04-05-2019
20
the operation abnormality (step S209), and ends the microphone array abnormal operation
determination process.
[0106]
When the microphone array abnormal operation determination unit 124 executes the
microphone array abnormal operation determination process as described above, the control unit
17 determines whether the microphone array abnormal operation determination unit 124
receives a notification such as a microphone array abnormal operation (see FIG. Step S104).
[0107]
When it is determined that there is no notification of the microphone array abnormal operation
and the like (step S104; No), the control unit 17 sets all the microphones 1211 to 1213 to be
used for the audio processing (step S105).
[0108]
On the other hand, when it is determined that there is a notification such as microphone array
abnormal operation (step S104; Yes), the control unit 17 performs audio processing so as not to
use the digital audio signal of the microphone performing abnormal operation for audio
processing. The microphones 1211 to 1213 to be used are changed (step S106).
[0109]
The control unit 17 controls the display unit 15 to display the microphone array abnormal
operation (step S107).
[0110]
The control unit 17 controls the voice output unit 13 to output a warning sound to notify the
user of the microphone array abnormal operation factor to the user of the voice processing
apparatus 1 (step S108).
[0111]
When the microphone setting (step S105) and the processing at the time of abnormal operation
determination (steps S106 to S108) are performed as described above, the control unit 17
instructs the voice processing unit 171 to perform voice processing (step S109). .
04-05-2019
21
[0112]
The audio processing unit 171 executes this audio processing in accordance with the flowchart
shown in FIG.
The echo canceller processing unit 1711 of the voice processing unit 171 performs echo
canceller processing (step S131).
[0113]
The noise removal processing unit 1712 performs noise removal processing (step S132).
The sound source localization processing unit 1713 performs sound source localization
processing (step S133).
The sound source separation processing unit 1714 performs sound source separation processing
(step S134).
The speech recognition processing unit 1715 performs speech recognition processing (step
S135).
When the speech processing unit 171 performs such speech processing, the speech processing
unit 171 ends the speech processing.
[0114]
When the voice processing unit 171 executes such voice processing, the control unit 17
determines whether to use the processed digital voice signal for communication based on the
input information supplied from the operation unit 14 ( S110).
[0115]
If it is determined that the digital audio signal after processing is to be used for communication
04-05-2019
22
(step S110; Yes), the control unit 17 supplies the digital audio signal after processing to the
communication unit 16 and causes the communication destination to transmit it (step S111).
[0116]
On the other hand, when it is determined that the digital audio signal after processing is not used
for communication (step S110; No), the control unit 17 determines whether the digital audio
signal is stored based on the input information supplied from the operation unit 14 Is
determined (step S112).
[0117]
When it is determined that the digital audio signal is to be stored (step S112; Yes), the control
unit 17 stores the processed digital audio signal in the information storage unit 11 (step S113).
[0118]
When the processed digital audio signal is transmitted to the communication destination (step
S111), when stored in the information storage unit 11 (step S113), or when it is determined not
to be stored (step S112; No), the control unit 17 It is determined based on the input information
supplied from the operation unit 14 whether to end the audio processing (step S114).
[0119]
The control unit 17 determines that the audio processing is not ended if the input information of
the end instruction of the audio processing is not supplied from the operation unit 14 (Step
S114; No).
[0120]
In this case, the control unit 17 instructs the voice acquisition unit 12 to execute the voice
acquisition process again (step S102).
[0121]
On the other hand, when the input information to the effect that the end command is instructed
is supplied from the operation unit 14 (step S114; Yes), the control unit 17 ends the voice
processing.
[0122]
04-05-2019
23
Next, the operation of the voice processing device 1 will be specifically described.
The voice acquisition unit 12 is instructed by the control unit 17 (processing of step S102) to
execute voice acquisition processing.
[0123]
The microphones 1211 to 1213 of the microphone array 121 receive voice and acquire analog
voice signals Sma1, Sma2, and Sma3 (processing of step S121).
[0124]
The A / D converter performs analog-digital conversion on the analog audio signals Sma1, Sma2
and Sma3 respectively output by the microphones 1211 to 1213, and outputs digital audio
signals Smd1, Smd2 and Smd3 (step S122). processing).
[0125]
The microphone dispersion correction unit 123 corrects the characteristic dispersion of the
microphones 1211 to 1213 using the microphone dispersion correction value 111 stored in the
information storage unit 11 (processing of step S123).
[0126]
When the voice acquisition unit 12 executes voice acquisition processing, the microphone array
abnormal operation determination unit 124 of the voice acquisition unit 12 is instructed by the
control unit 17 (processing of step S103) to execute microphone array abnormal operation
determination processing.
[0127]
The silence determination unit 1241 of the microphone array abnormal operation determination
unit 124 performs silence determination (processing of steps S201 to S203).
[0128]
As shown in FIG. 7, when all the signal levels Lmd_1, Lmd_2, and Lmd_3 of the digital audio
signals Smd1, Smd2, and Smd3 are less than the prescribed level Lmd_th at time t1, the silence
04-05-2019
24
determination unit 1241 measures time. (Process of step S202).
[0129]
When all the signal levels Lmd_1, Lmd_2, and Lmd_3 are less than the prescribed level Lmd_th
even at the time t2 when the prescribed time Tth is exceeded from the time t1 (step S203; Yes),
the silence determination unit 1241 determines silence. judge.
[0130]
The microphone array abnormal operation determination unit 124 does not perform the
microphone array abnormal operation determination when the silence determination unit 1241
determines that there is no sound.
[0131]
Next, when voices are input to the microphones 1211 to 1213 and the microphones 1211 to
1213 operate normally, as shown in FIG. 8A, the signal levels Lmd_1, Lmd_2, and Lmd_3 are all
equal to or higher than the prescribed level Lmd_th. It becomes.
In this case, the silence determination unit 1241 determines that the voice is not silent (step
S201; No).
[0132]
When the silence determination unit 1241 determines that the signal is not silent, the volume
comparison unit 1242 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 (processing in step
S204).
[0133]
As a result of comparison, the volume comparison unit 1242 compares the signal level difference
ΔL_12 between the signal levels Lmd_1 and Lmd_2, the signal level difference ΔL_23 between
the signal levels Lmd_2 and Lmd_3, and the signal levels Lmd_3 and Lmd_1 as shown in FIG. It is
assumed that the signal level difference ΔL_31 is acquired.
[0134]
04-05-2019
25
When the standard value ΔL_th of the signal level differences ΔL_12, ΔL_23, ΔL_31 is set in
advance as shown in FIG. 8B, the signal level differences ΔL_12, ΔL_23, ΔL_31 all become the
standard value ΔL_th or more.
[0135]
Further, as shown in FIG. 8C, the correlation coefficient calculation unit 1243 is the correlation
coefficient r12 between the signal levels Lmd_1 and Lmd_2, the correlation coefficient r23
between the signal levels Lmd_2 and Lmd_3, and the signal levels Lmd_3 and Lmd_1. It is
assumed that the correlation coefficient r31 of is calculated (processing of step S205).
[0136]
When the standard value r_th of the correlation coefficients r12, r23, r31 is set in advance as
shown in FIG. 8B, the correlation coefficients r12, r23, r31 all become the standard value r_th or
more.
[0137]
In the microphone array abnormal operation judging unit 124, the signal level differences
ΔL_12, ΔL_23, ΔL_31 are all equal to or larger than the standard value ΔL_th, and the
correlation coefficients r12, r23, r31 are all equal to or larger than the standard value r_th.
Determines that the microphones 1211 to 1213 are operating normally (step S206; No).
[0138]
When the microphone array abnormal operation determination unit 124 determines as described
above and notifies the control unit 17 of no microphone array abnormal operation (the process
of step S207; step S104; No), the voice processing unit 171 performs all the processes. The voice
processing process is performed using the microphones 1211 to 1213 (processes of steps S105
and S109).
[0139]
Next, when a failure occurs in the microphone 1212 or there is a microphone closure, as shown
in FIG. 9A, the signal level Lmd_2 of the digital audio signal Smd2 becomes lower than the
prescribed level Lmd_th.
[0140]
However, since the signal levels Lmd_1 and Lmd_3 of the digital audio signals Smd1 and Smd3
04-05-2019
26
are equal to or higher than the predetermined level Lmd_th, the silence determination unit 1241
determines that the signal is not silence (Step S201; No).
[0141]
When the silence determination unit 1241 determines that the signal is not silent, the volume
comparison unit 1242 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 (processing in step
S204).
As a result of comparison, it is assumed that the sound volume comparison unit 1242 has
acquired the signal level differences ΔL_12, ΔL_23, and ΔL_31 as shown in FIG. 9B.
[0142]
When the standard value ΔL_th is preset as shown in FIG. 9B, the signal level difference ΔL_31
is smaller than the standard value ΔL_th, but the signal level differences ΔL_12 and ΔL_23 are
larger than the standard value ΔL_th.
[0143]
Further, it is assumed that the correlation coefficient calculation unit 1243 calculates the
correlation coefficients r12, r23, r31 as shown in FIG. 9C (processing of step S205).
[0144]
When the standard value r_th is preset as shown in FIG. 9B, the correlation coefficient r31 is
higher than the standard value r_th, but the correlation coefficients r12 and r23 are lower than
the standard value r_th.
[0145]
If at least one of the signal level differences ΔL_12, ΔL_23, ΔL_12 is equal to or greater than
the standard value ΔL_th, and one of the correlation coefficients r12, r23, r31 is less than the
standard value r_th, the microphone array abnormal operation determination unit 124 It is
determined that the microphone array is in an abnormal operation (Step S206; Yes).
[0146]
04-05-2019
27
When it is determined that the microphone array abnormal operation is performed, the
microphone array abnormal operation determination unit 124 determines the microphone of the
operation abnormal (processing of step S208).
[0147]
The microphone array abnormal operation determination unit 124 determines the highest signal
level.
As shown in FIG. 9A, the highest signal level is the signal level Lmd_3.
[0148]
The microphone array abnormal operation determination unit 124 examines the signal level
difference which is equal to or greater than the standard value ΔL_th with reference to the
signal level Lmd_3.
When the calculation result is as shown in FIG. 9B, the signal level difference ΔL_23 becomes
equal to or larger than the standard value ΔL_th based on the signal level Lmd_3.
[0149]
Further, the microphone array abnormal operation determination unit 124 examines the
correlation coefficient which is equal to or greater than the standard value r_th.
When the calculation result is as shown in FIG. 9C, the correlation coefficients r12 and r23 are
less than the standard value r_th.
[0150]
When the microphone array abnormal operation determination unit 124 determines that the
signal level difference ΔL_23 is equal to or larger than the standard value ΔL_th and the
04-05-2019
28
correlation coefficients r12 and r23 are smaller than the standard value r_th, the microphone
1212 corresponding to the signal level Lmd_2 is abnormal in operation. Determined as a
microphone.
[0151]
When the microphone array abnormal operation determination unit 124 determines in this
manner, the microphone array abnormal operation determination unit 124 notifies the control
unit 17 of the microphone array abnormal operation (processing of step S209).
[0152]
In the case of the microphone array abnormal operation, it is not possible to perform normal
sound processing using all the microphones 1211 to 1213.
[0153]
Therefore, when notified by the microphone array abnormal operation determination unit 124
(step S104; Yes), the control unit 17 changes the microphones 1211 to 1213 used for the audio
processing (processing of step S106).
[0154]
As described above, by changing the microphones 1211 to 1213 used for audio processing, the
control unit 17 prevents the analog audio signal Sma2 input by the microphone 1212 in
abnormal operation from being used for audio processing.
[0155]
The display unit 15 displays the abnormal operation of the microphone array 121 (processing of
step S107), and the voice output unit 13 outputs a warning sound (processing of step S108).
[0156]
Next, when a failure occurs in the microphone 1211 and the microphone 1213 or a microphone
closing occurs, as shown in FIG. 10A, the signal level Lmd_2 of the digital audio signal Smd2
becomes lower than the prescribed level L_th.
[0157]
However, since the signal level Lmd_2 of the digital audio signal Smd2 is equal to or higher than
04-05-2019
29
the predetermined level Lmd_th, the silence determination unit 1241 determines that the signal
is not silent (Step S201; No).
[0158]
When the silence determination unit 1241 determines that the signal is not silent, the volume
comparison unit 1242 compares the signal levels Lmd_1, Lmd_2, and Lmd_3 (processing in step
S204).
As a result of comparison, it is assumed that the sound volume comparison unit 1242 has
acquired the signal level differences ΔL_12, ΔL_23, and ΔL_31 as shown in FIG.
[0159]
When the standard value ΔL_th is set in advance as shown in FIG. 10B, the signal level
difference ΔL_31 becomes smaller than the standard value ΔL_th as in the case of FIG. 9B, but
the signal level differences ΔL_12, ΔL_23. Is equal to or greater than the standard value ΔL_th.
[0160]
Further, it is assumed that the correlation coefficient calculation unit 1243 calculates the
correlation coefficients r12, r23, r31 as shown in FIG. 10C (processing of step S205).
[0161]
When the standard value r_th is set in advance as shown in FIG. 10C, the correlation coefficients
r11, r12 and r31 are smaller than the standard value r_th.
[0162]
Among the signal level differences ΔL_12, ΔL_23, and ΔL_12, the microphone array abnormal
operation determination unit 124 has the signal level differences ΔL_12 and ΔL_23 equal to or
greater than the standard value ΔL_th and all the correlation coefficients r12, r23, and r31 less
than the standard value r_th. It is determined that the microphone array is in an abnormal
operation (Step S206; Yes).
04-05-2019
30
[0163]
When it is determined that the microphone array abnormal operation is performed, the
microphone array abnormal operation determination unit 124 determines the microphone of the
operation abnormal (processing of step S208).
[0164]
As shown in FIG. 10A, since the highest signal level is the signal level Lmd_2, the microphone
array abnormal operation determination unit 124 examines the signal level difference with
reference to the signal level Lmd_3, and the standard value ΔL_th is obtained. It is determined
that the above signal level difference is the signal level differences ΔL_12 and ΔL_23.
[0165]
When the microphone array abnormal operation judging unit 124 judges that the signal level
differences ΔL_12 and ΔL_23 become equal to or larger than the standard value ΔL_th and all
the correlation coefficients r12, r23 and r31 become smaller than the standard value r_th, the
microphones 1211 and 1213 operate abnormally And determine the microphone.
[0166]
When the microphone array abnormal operation determination unit 124 determines in this
manner, the microphone array abnormal operation determination unit 124 notifies the control
unit 17 of the microphone array abnormal operation (processing of step S209).
[0167]
When notified by the microphone array abnormal operation judging unit 124 (step S104; Yes),
the control unit 17 does not use the analog audio signals Sma1 and Sma3 input by the
microphones 121 and 121 which are operating abnormally for the audio processing. (Process of
step S106).
[0168]
As a result of the audio processing unit 171 performing audio processing, when communication
is performed according to the user's operation (step S110; Yes), the communication unit 16
transmits the processed audio signal.
[0169]
When the processed audio signal is stored (step S112; Yes), the control unit 17 stores the
04-05-2019
31
processed audio signal in the information storage unit 11 (processing of step S113).
[0170]
As described above, according to the present embodiment, the microphone array abnormal
operation determination unit 124 determines the microphone array based on the signal level
difference between the signal levels of the digital audio signals corresponding to the
microphones 1211 to 1213 and the correlation coefficient. It was made to judge abnormal
operation.
[0171]
Therefore, the abnormal operation determination of the three microphones 1211 to 1213
configuring the microphone array 121 can be performed without using a signal having a power
equal to or higher than a predetermined level. Judgment can be made.
[0172]
Further, when the microphone array abnormal operation determination unit 124 notifies of the
microphone array abnormal operation etc., the control unit 17 uses the microphones 1211 to
1213 for audio processing so as not to use the microphone of the abnormal operation for audio
processing. I made a change of.
[0173]
Therefore, even during voice processing, normal voice signal processing can be continued.
[0174]
Further, the silence determination unit 1241 performs silence determination, and the
microphone array abnormal operation determination unit 124 does not perform the microphone
array abnormal operation determination process when the silence determination unit 1241
determines that the sound is silent.
[0175]
Therefore, even when a voice signal in which a silent section exists is selected as an external
input signal for failure diagnosis, in this silent section, all microphones 1211 to 1213 are not
determined to be faulty, so the accuracy of the microphone array abnormal operation
determination Can be enhanced.
04-05-2019
32
[0176]
Further, the microphone array abnormal operation determination unit 124 executes microphone
array abnormal operation determination processing during audio processing.
Therefore, for example, when the microphone arrays 1211 to 1213 are mounted on a built-in
device such as a mobile phone, even if the microphone can not be used temporarily due to the
microphone blocking, if the microphone blocking is released, it is used again. Will be able to
[0177]
Note that various forms can be considered in practicing the present invention, and the present
invention is not limited to the above embodiment.
[0178]
That is, in the present embodiment, only a part of the above-described principle may be used, and
the present invention is not limited to the above-described embodiment, and the above-described
principle is combined to form a new embodiment. can do.
[0179]
For example, in the above embodiment, the case where the microphone array 121 is configured
by three microphones 1211 to 1213 has been described.
However, the number of microphone arrays 121 is not limited to three, and may be any number
of two or more.
[0180]
In the above embodiment, the confirmation sound output from the voice output unit 13 at the
start of the voice processing has been described as being used to notify the user of the voice
processing apparatus 1 (step S101).
04-05-2019
33
[0181]
However, this confirmation sound can also be used for microphone array abnormal operation
detection processing.
That is, if voices such as specified volume and frequency are input to the microphones 1211 to
1213 and silence determination is performed, it is possible to determine whether the operation
of the microphone array 121 is correct or not, and the microphone array abnormal operation
determination unit 124 It is possible to determine the microphone array abnormal operation
with high accuracy.
[0182]
As described above, if the confirmation sound is used in the microphone array abnormal
operation determination process, the confirmation sound becomes more effective.
[0183]
In addition, the sound processing unit 171 can initialize parameters to be used by setting in
advance the volume, frequency, and the like of the confirmation sound.
[0184]
The echo canceller processing unit 1711 can also suppress the confirmation sound component
by the sound acquisition unit 12 acquiring the confirmation sound component.
[0185]
Also, the warning sound output from the voice output unit 13 in step S109 can be used for the
microphone array abnormal operation detection process, as with the confirmation sound.
[0186]
In the above embodiment, the microphone array abnormal operation judging unit 124 is
instructed by the control unit 17 to execute the microphone array abnormal operation judging
process.
04-05-2019
34
However, the microphone array abnormal operation determination unit 124 can also execute the
microphone array abnormal operation determination process without being instructed by the
control unit 17.
[0187]
That is, the microphone array abnormal operation determination unit 124 may execute the
microphone array abnormal operation determination process, for example, once every two
seconds, every time a predetermined time set in advance elapses.
[0188]
In this case, the microphone array abnormal operation determination unit 124 stores the
determination result of the microphone array abnormal operation in the information storage unit
11 each time this process is executed, and the control unit 17 determines with reference to the
information storage unit 11. Obtain the already determined result.
[0189]
With such a configuration, the processing amount of the control unit 11 and the like can be
reduced, and power consumption can be reduced.
In addition, even when the speech processing device 1 has low processing capacity, speech
processing can be performed in real time.
[0190]
In the above embodiment, when the control unit 17 sets the microphones 1211 to 1213 (step
S106), all the microphones 1211 to 1213 are set to be used for audio processing.
[0191]
However, the control unit 17 may set, for example, not all of the microphones used for audio
processing to the microphone 1211 and the microphone 1212 or the microphone 1211 and the
microphone 1213.
04-05-2019
35
[0192]
In addition, when the microphones 121 and 121 set the microphones used for audio processing
to the control unit 17 and the microphone array abnormal operation judging unit 124 is notified
of the determination result of the microphone array abnormal operation, the control unit 17
performs the setting. The microphones can also be changed to the microphones 121 and 1213
(step S106).
[0193]
In the above embodiment, the display unit 15 is configured to display the microphone array
abnormal operation when notified of the microphone array abnormal operation or the like.
However, the display is not limited to this. For example, the display unit 15 may display a
message prompting the user of the voice processing device 1 to confirm whether the microphone
blocking has occurred.
[0194]
In the above embodiment, if the silence determination unit 1241 determines that there is no
sound, it can not be determined whether the microphone array abnormal operation or the voice
is not input to the microphones 1211-1213. It is also possible not to perform voice processing.
[0195]
In this case, the microphone array abnormal operation determination unit 124 notifies the
control unit 17 that there is no sound when the silence determination unit 1241 determines that
there is no sound, and the control unit 17 receives this notification. End the audio processing.
[0196]
In the above embodiment, the microphone array abnormal operation judging unit 124 judges the
microphone array abnormal operation based on the calculation result of the volume comparing
unit 1242 and the calculation result of the correlation coefficient calculating unit 1243.
[0197]
However, the microphone array abnormal operation judging unit 124 judges the microphone
array abnormal operation using either one of the calculation result of the volume comparing unit
04-05-2019
36
1242 and the calculation result of the correlation coefficient calculating unit 1243. It is also
good.
[0198]
In addition, the microphone array abnormal operation determination unit 124 may perform the
microphone array abnormal operation determination using a method other than the calculation
result of the sound volume comparison unit 1242 and the calculation result of the correlation
coefficient calculation unit 1243.
[0199]
Details of this process will be described later.
Note that this process may be performed only once every predetermined time period.
For example, this processing may be performed only once every two seconds, and when this
processing is not performed, it may be regarded as the same as the previous processing result.
By doing so, it is possible to reduce the processing amount of the speech processing device 1 for
implementing the present invention and to realize low power consumption.
[0200]
Further, in the above embodiment, the programs are described as being stored in advance in the
memory or the like.
However, a program for operating the audio processing device as all or part of the device or
executing the above-mentioned processing may be a flexible disk, a compact disk read-only
memory (CD-ROM), a digital versatile disk (DVD) May be stored in a computer-readable recording
medium such as MO (Magneto Optical disk) and distributed, installed on another computer,
operated as the means described above, or executed the steps described above .
04-05-2019
37
[0201]
Furthermore, the program may be stored in a disk device or the like of a server device on the
Internet, for example, superimposed on a carrier wave and downloaded to a computer.
[0202]
It is a block diagram showing the composition of the speech processing unit concerning the
embodiment of the present invention.
It is a flowchart (the 1) which shows the audio processing which the control part shown in FIG. 1
performs.
It is a flowchart (the 2) which shows the audio processing which the control part shown in FIG. 1
performs.
It is a flowchart which shows the audio | voice acquisition process which the audio | voice
acquisition part shown in FIG. 1 performs.
It is a flowchart which shows the microphone array abnormal operation | movement
determination processing which the microphone array abnormal operation | movement
determination part shown in FIG. 1 performs.
It is a flowchart which shows the audio processing process which the audio processing part
shown in FIG. 1 performs.
It is a figure which shows a concrete operation | movement in case an audio processing
apparatus performs silence determination.
It is a figure which shows the concrete operation | movement of the microphone array abnormal
operation | movement determination processing which a microphone array abnormal operation |
movement determination part performs, when all the microphones are operate | moving
04-05-2019
38
normally, (a) is a signal level of each microphone The relationship with the preset standard value
is shown, (b) shows the relationship between the signal level difference between each signal
level, and the preset standard value, (c) shows the phase relationship between each signal level
The relationship between the number and the preset standard value is shown.
It is a figure which shows the concrete operation | movement of the microphone array abnormal
operation | movement determination processing which a microphone array abnormal operation |
movement determination part performs, when failure etc. generate | occur | produce in one
microphone, (a) is a signal level of each microphone The relationship with the set standard value
is shown, (b) shows the relationship between the signal level difference between each signal
level, and the preset standard value, (c) is a correlation coefficient between each signal level And
the preset standard value.
It is a figure which shows the concrete operation | movement of the microphone array abnormal
operation | movement determination processing which a microphone array abnormal operation |
movement determination part performs, when a failure etc. generate | occur | produces in two
microphones. The relationship with the set standard value is shown, (b) shows the relationship
between the signal level difference between each signal level, and the preset standard value, (c) is
a correlation coefficient between each signal level And the preset standard value.
Explanation of sign
[0203]
1 voice processing device 11 information storage unit 12 voice acquisition unit 17 control unit
121 microphone array 124 microphone array abnormal operation determination unit 171 audio
processing units 1211 to 1213 microphone 1241 silence determination unit 1242 volume
comparison unit 1243 correlation coefficient calculation unit
04-05-2019
39
Документ
Категория
Без категории
Просмотров
0
Размер файла
49 Кб
Теги
jp2009100072
1/--страниц
Пожаловаться на содержимое документа