close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JP2005109533

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 JP2005109533
PROBLEM TO BE SOLVED: To accurately determine and suppress howling. A frequency analyzer
analyzes a frequency component of an input signal from a microphone, a level detector detects
an input signal level, and a designated frequency component of the input signal is attenuated and
output. A control unit 34 controls the frequency characteristics of the BEF 50 and the BEF 50.
When the control unit 34 determines that the peak in the frequency spectrum is equal to or
higher than the threshold, it controls to attenuate the component of the frequency fa of the peak,
and when it is determined that the level of the input signal decreases, the frequency While
maintaining the attenuation of the component, when it is determined that the component does
not decrease, control is performed so as to cancel the attenuation of the frequency component of
the peak. [Selected figure] Figure 1
Howling suppression device
[0001]
The present invention relates to a technique for suppressing howling when inputting a song, for
example, a karaoke apparatus.
[0002]
In audio equipment that amplifies and then outputs a signal input from a microphone and the
like from a speaker, a closed loop is formed by reinputting the output from the speaker to the
microphone, which may cause unpleasant and unpleasant howling. .
15-04-2019
1
Since this howling occurs in the input signal of the microphone when a large constant frequency
component with a peak continues for a long time, various techniques for suppressing howling
have been proposed focusing on this feature. There is. As such a technique, for example, when
the microphone input signal is subjected to frequency analysis, and the above-mentioned feature
appears in the analysis result, it is determined that howling has occurred, and the frequency of
the notch filter is set to the peak frequency. Patent Document 1 discloses a technique for
attenuating howling frequency components by filtering an input signal of
[0003]
JP-A-7-143 034 (see FIGS. 1 and 3 and paragraphs 0026 and 0030)
[0004]
By the way, according to the above-mentioned technology, when voices such as lectures and daily
conversations are input with a microphone, it is relatively accurate to determine whether or not
howling has occurred, but when voices such as singing are input The following problems have
been pointed out: the judgment is likely to be inaccurate.
That is, depending on how to sing, the peak of the voice may be large and the constant frequency
component may continue for a long time, and the result of analysis of the input signal of the
microphone at this time is very close to the howling feature. The present invention has been
made in view of the above-described circumstances, and a purpose thereof is a howling
suppression device capable of appropriately suppressing howling even when a voice such as
singing is input. It is to provide.
[0005]
In order to achieve the above object, a howling suppression apparatus according to the present
invention comprises: analysis means for analyzing frequency components of an input signal from
a microphone; detection means for detecting the level of the input signal; and filter for filtering
the input signal And controlling the frequency characteristic of the filter so as to attenuate the
frequency component of the peak when the peak in the frequency spectrum determined by the
analysis means is equal to or greater than the threshold, and reducing the level of the input
signal When it is determined that the frequency component has decreased, the attenuation of the
15-04-2019
2
frequency component is maintained, and when it is determined that the frequency component
has not decreased, the filter is controlled to cancel the attenuation of the frequency component
of the peak. And control means. According to the present invention, when the magnitude of the
peak frequency is equal to or greater than the threshold value, the peak frequency component of
the input signal is controlled to be attenuated once, and then whether or not the level of the
input signal decreases. Determine. Here, when the level of the input signal decreases, the peak
frequency component is maintained to be attenuated, assuming that the cause of the level of the
input signal being equal to or higher than the threshold earlier is howling, so the howling is
appropriately suppressed. On the other hand, when the level of the input signal does not
decrease, the above-mentioned cause is the cancellation of the peak frequency component except
for howling, so that the inappropriate howling suppression is prevented.
[0006]
In the present invention, after the control means controls the filter to cancel the attenuation of
the peak frequency component, the peak frequency is detected even if the level of the input
signal detected by the detection means is equal to or higher than the threshold value. A
configuration that inhibits the re-attenuation of the component is preferred. According to this
configuration, even if a similar phenomenon occurs again, the attenuation of the peak frequency
component is prohibited, so that inappropriate howling suppression is prevented in advance.
[0007]
Hereinafter, embodiments of the present invention will be described with reference to the
drawings. FIG. 1 is a block diagram showing the configuration of a karaoke apparatus to which
the howling suppression apparatus according to the present embodiment is applied. In this
figure, the microphone 10 converts the singer's voice into an analog singing signal. The
frequency analyzer 20 comprises an A / D converter 22, a discrete Fourier transformer (DFT) 24
and an integrator 26. Among these, the A / D converter 22 samples and sings a singing signal
from the microphone 10 at a sampling frequency of 32 kHz, for example, and outputs the data
by quantizing the same. The DFT 24 Fourier-transforms the data of the quantized singing signal
to obtain its frequency spectrum. The integrator 26 integrates (averages) data indicating the size
of each frequency band in this frequency spectrum, for example, every 2 seconds, that is, every
period of a frequency sufficiently lower than the sampling frequency of the singing signal.
[0008]
15-04-2019
3
The singing signal from the microphone 10 is also supplied to the level detector 32 and the BEF
50. Among these, the level detector 32 detects the level (volume) of the singing signal. A BEF
(Band Eliminate Filter) 50 is a band elimination filter that attenuates only components near the
notch frequency fn by a designated attenuation b when the notch frequency fn is set by the
control unit 34 described later. The control unit 34 controls the frequency characteristic of the
BEF 50 by means of a programmed microcomputer or the like.
[0009]
On the other hand, the storage unit 40 stores MIDI data for a plurality of songs. The performance
sequencer 42 reads out and outputs MIDI data of the music designated by the operation unit (not
shown) from the storage unit 40 in accordance with the passage of time from the start of the
performance. The tone generator 44 generates and outputs a musical tone signal based on the
performance data supplied from the performance sequencer 42. The mixer 52 mixes the singing
signal filtered by the BEF 50 with the musical tone signal by the sound source 44. The sound
system 54 emits the mixed signal after appropriate amplification.
[0010]
Next, the operation of the karaoke apparatus having such a configuration will be described. FIG.
2 is a flowchart showing howling suppression operation by the karaoke apparatus. First, the
control unit 34 executes an initialization process (step Sa11). The initialization process includes a
process of flattening the BEF 50 over the entire frequency range as shown in FIG. 4A, and a
process of clearing the suppression prohibition frequency fs and various memory states
described later. The performance sequencer 42 reads out MIDI data of a designated song from
the storage unit 40 as time passes. Since the read MIDI data is supplied to the sound source 44, a
musical tone signal which is an accompaniment sound of karaoke is output.
[0011]
On the other hand, when the singer starts singing along with the output of the musical tone
signal, the voice is converted into a singing signal by the microphone 10 and supplied to the A /
D converter 22, the level detector 32, and the BEF 50, respectively. Here, since the frequency
15-04-2019
4
characteristic of the BEF 50 is flattened by the initialization process, immediately after the
operation starts, the singing signal by the microphone 10 is mixed with the musical tone signal
as it is and emitted. In addition, the singing signal is converted into digital data by the A / D
converter 22, and the frequency spectrum is obtained by the DFT 24. Data indicating the size of
each frequency range in this frequency spectrum is temporally averaged by integration by the
integrator 26, respectively.
[0012]
The control unit 34 determines whether the frequency spectrum averaged by the integrator 26
includes a portion greater than or equal to the threshold th (step Sa12). As shown in FIG. 3A, if
there is no component above the threshold th in the frequency spectrum, the control unit 34
determines that howling has not occurred, and returns the processing procedure to step Sa12. .
For this reason, as long as the frequency spectrum continues to fall below the threshold th, in the
present embodiment, the subsequent processing is not executed at all. On the other hand, as
shown in FIG. 3 (b), when the frequency spectrum has a threshold value th or more, it means that
a constant frequency component with a large peak continues for a long time, so howling occurs.
There is a possibility. However, since it is also considered that the frequency spectrum has
become equal to or higher than the threshold th due to causes other than howling, in order to
confirm this point, the control unit 34 continues the processing of the following steps Sa13 to
Sa18.
[0013]
First, the control unit 34 specifies the peak frequency fa which is equal to or higher than the
threshold th in the frequency spectrum (step Sa13), and determines whether the peak frequency
fa does not match the frequency set as the suppression prohibition frequency fs. It discriminates
(step Sa14). If the peak frequency fa corresponds to the suppression prohibition frequency fs, as
described later, it is already known that the cause of the peak frequency fa becoming equal to or
higher than the threshold th is other than howling, the control unit 34 The processing procedure
returns to step Sa12. For this reason, the howling suppression processing for the peak frequency
fa is not executed thereafter. On the other hand, when the peak frequency fa does not match the
frequency set as the suppression prohibition frequency fs or when the suppression prohibition
frequency fs is not set, the control unit 34 sets the peak frequency fa as the notch frequency fn
of the BEF 50. While setting, the attenuation bn of the frequency band is set to b1. Therefore, as
shown in FIG. 4B, the BEF 50 has a frequency characteristic such that a component near the peak
frequency fa is attenuated at the attenuation b1.
15-04-2019
5
[0014]
Here, if howling occurs before the frequency characteristic of BEF 50 is set, after setting, the
component of peak frequency fa of the singing signal by microphone 10 is attenuated and mixed
with the musical tone signal to be emitted. Therefore, even if the microphone 10 reinputs the
sound emission, the howling is suppressed. Thus, the level of the singing signal by the
microphone 10 should be reduced. On the other hand, if howling does not occur before the
frequency characteristic of BEF 50 is set, the component near peak frequency fa will be
attenuated by BEF 50 after setting, but the level of the singing signal should not decrease as a
whole It is.
[0015]
Therefore, the control unit 34 determines whether the level of the singing signal has decreased
before or after setting fa as the notch frequency fn in the BEF 50 using the detection result of the
level detector 32 (step Sa16). . If the control unit 34 determines that the level of the singing
signal decreases and howling is suppressed, the control unit 34 returns the processing procedure
to step Sa12. Therefore, the BEF 50 is maintained in the state of the frequency characteristic that
suppresses howling, and the same operation is performed even when a portion that becomes the
threshold th is newly generated in the frequency spectrum of the singing signal. Therefore, in the
present embodiment, the notch frequency fn may be set at a plurality of positions in the BEF 50.
[0016]
On the other hand, if the level of the singing signal does not decrease before and after setting fa
as the notch frequency fn in the BEF 50, the control unit 34 determines that the cause of the
frequency spectrum becoming the threshold th or more is other than howling. Then, the
attenuation degree bn changed in step Sa15 is returned to the original state, and processing for
setting the attenuation degree of the peak frequency fa to zero is executed (step Sa17). This
process is executed when the howling does not occur, so if the attenuation degree is sharply
returned, the listener may feel the change and give a sense of discomfort or the like. Therefore,
the degree of attenuation of the peak frequency fa may be gradually returned gradually with time
as shown in FIGS. 4 (b), 4 (c) and 4 (d). Furthermore, the control unit 34 sets the peak frequency
fa which has not been related to the occurrence of howling to the suppression prohibition
15-04-2019
6
frequency fs (step Sa18), and thereafter returns the processing procedure to step Sa12.
Therefore, after that, even if the frequency spectrum of the singing signal becomes equal to or
higher than the threshold th, if the peak frequency is fa, the determination result in step Sa14
becomes “No”. Processing will be prohibited. Therefore, the BEF 50 is not reset to the
characteristic of attenuating components near the peak frequency fa.
[0017]
As described above, in the present embodiment, the frequency characteristic of the BEF 50 is
such that the peak frequency fa that is equal to or higher than the threshold th is attenuated
when a portion that becomes the threshold th occurs in the frequency spectrum of the singing
signal input by the microphone 10 Is set. Here, if the level of the singing signal decreases from
the front to the back of the set, the howling that has occurred is suppressed, so the frequency
characteristics of the BEF 50 are maintained while the level of the singing signal decreases.
Otherwise, since howling does not occur originally, the frequency characteristic of the BEF 50 is
restored, and thereafter, the frequency characteristic is prohibited from being set so as to
attenuate the peak frequency fa. Therefore, according to the present embodiment, it is possible to
appropriately suppress howling even when a voice such as singing is input.
[0018]
In the embodiment, if the level of the singing signal decreases from before to after the frequency
characteristic of BEF 50 is set, it is determined that howling has occurred, and the frequency
characteristic is such that BEF 50 attenuates peak frequency fa. However, while howling occurs
when a specific condition is satisfied, the condition may be resolved if time passes to a certain
extent. Therefore, when a predetermined time has elapsed, the control unit 34 preferably follows
the attenuation degree as shown in FIGS. 4 (b), 4 (c) and 4 (d). It may be configured to gradually
return. When the degree of attenuation is returned, there may be a case where the howling
occurrence condition is not resolved, but in that case, the determination result in step Sa12
becomes “Yes”, and the peak frequency fa is again set to BEF 50. The frequency characteristic
to be attenuated is set.
[0019]
Further, in the embodiment, since the notch frequency fn may be set in a plurality of places in
15-04-2019
7
the BEF 50, an upper limit may be set to the set number. When the set number reaches the upper
limit, the control unit 34 may be configured to delete the notch frequency in order from the
oldest time.
[0020]
The control unit 34 may be configured to set the frequency characteristic of the BEF 50 and
store characteristics such as the frequency spectrum before setting the frequency characteristic
and the level of the singing signal when the level of the singing signal decreases. Then, when a
portion to be the threshold th occurs in the frequency spectrum of the singing signal input by the
microphone 10, if the stored feature is satisfied, “Yes” is determined in step Sa12, but if the
feature is not satisfied, It may be determined as "No".
[0021]
In the embodiment, the BEF 50 is an analog filter, but may be a digital filter. When BEF is a
digital filter, the output of A / D converter 22 is also supplied to BEF, and a D / A converter is
inserted in the previous stage of sound system 54 to convert it back into an analog signal. good.
Also, in this case, the output of the sound source 44 may be digital data, and the mixer 52 may
perform addition of the digital data.
[0022]
In the embodiment, when the control unit 34 determines “Yes” in step Sa12, the peak
frequency fa is set to the notch frequency fn using the BEF 50, but in order to attenuate
uniformly over the entire frequency range, in other words If so, the control may be made to
simply lower the gain. That is, when the determination result in step Sa12 is "Yes", it may be
attenuated so as to include at least the component of the peak frequency fa. However, in the
configuration in which the gain is reduced, the sound level of the output sound drops if the cause
of the frequency spectrum including ones above the threshold th is other than howling, which
may make the listener uncomfortable. There is sex. Further, although the BEF 50 is disposed at
the front stage of the mixer 52, it may be disposed at the rear stage of the mixer 52.
[0023]
15-04-2019
8
It is a block diagram which shows the structure of the karaoke apparatus to which the howling
suppression apparatus which concerns on embodiment of this invention is applied. It is a
flowchart which shows the suppression operation of howling in the karaoke apparatus. It is a
figure which shows extraction of the peak frequency in the karaoke apparatus. It is a figure
which shows the frequency characteristic of BEF in the karaoke apparatus.
Explanation of sign
[0024]
DESCRIPTION OF SYMBOLS 10 ... Microphone, 20 ... Frequency analyzer, 32 ... Level detector, 34
... Control part, 42 ... Performance sequencer, 44 ... Sound source, 46 ... Converter, 50 ... BEF, 52 ...
Mixer, 54 ... Sound system.
15-04-2019
9
Документ
Категория
Без категории
Просмотров
0
Размер файла
19 Кб
Теги
jp2005109533, description
1/--страниц
Пожаловаться на содержимое документа