close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPH08275279

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 JPH08275279
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a
voice pickup system, and more particularly to a voice pickup system for detecting bone
conduction sounds and airway sounds with a pickup.
[0002]
2. Description of the Related Art A bone conduction sound pickup for detecting bone conduction
sound is known as a microphone used in an environment where there is noise.
[0003]
This bone conduction sound pickup is in close contact with the user's face or head, and detects
the vibration of the face or head accompanying the user's voice.
In addition, as a similar microphone, there is also a microphone that detects an airway sound of
the ear canal by inserting a pickup having a shape like an earphone into the outer ear.
[0004]
The pickup for detecting bone conduction sound and airway sound as described above has an
advantage that it is easy to detect a target user's voice without relatively picking up ambient
10-04-2019
1
noise.
[0005]
However, when the spectrum of normal speech is 300 Hz to 3 kHz, there is a problem that the
detection level decreases in the frequency region of 1 kHz or more as shown in the characteristic
B of FIG. doing.
[0006]
That is, the detection sound of the pickup in the form of indirectly detecting the sound, such as
bone conduction sound and airway sound, emphasizes the low frequency component of the
sound in comparison with the sound signal detected by the ordinary microphone (characteristic
A in FIG. 5) As a result, it seems that the high frequency component is gradually reduced
(characteristic B in FIG. 5), and there is a problem that the clarity is reduced.
[0007]
FIG. 6 is a characteristic diagram showing frequency characteristics when a male voice is actually
detected by a bone conduction sound pickup.
Also from this characteristic diagram, it is possible to read how the high frequency component is
reduced.
Actually, the degree of reduction of the high frequency component is slightly different depending
on the pressure when pressing the bone conduction sound pickup to the face or head, the
distinction between male and female, etc. Absent.
[0008]
The present invention has been made in view of the above problems, and an object thereof is to
provide a voice pickup system capable of transmitting voice in a state of high clarity without
picking up ambient noise. It is in.
[0009]
SUMMARY OF THE INVENTION As a result of intensive research aimed at improving the
10-04-2019
2
problems such as the intelligibility and the like expected in the conventional voice pickup system,
the inventor of the present application has conventionally made the intelligibility in terms of
frequency characteristics. The present invention has been accomplished by finding a
configuration capable of transmitting voice with high clarity even in a pickup for detecting bone
conduction sound and airway sound, which has had a problem in the above.
[0010]
Accordingly, the present invention, which is a means for solving the problems, is configured as
described below.
(1) That is, a first means for solving the above-mentioned problems is a voice pickup system for
detecting bone conduction sound or airway sound by a pickup, wherein the voice recognition
means capable of recognizing a sound detected by the pickup; Spectrum generation means for
generating a spectrum of frequency components when data of the recognized sound is detected
by a normal microphone using data of the sound recognized by the recognition means; a
spectrum generated by the spectrum generation means; A comparison means for comparing
missing spectrum components by comparing the spectrum of the sound detected by the pickup,
and a sound of the frequency components missing from the sound detected by the pickup from
the comparison result of the comparing means Missing spectrum generating means for
generating a complementary sound, sound detected by the pickup and the missing spectrum
Synthesizing means for outputting by synthesizing the generated by generating means
complementary sound, a sound pickup system characterized by having a.
[0011]
In such a voice pickup system, the complementary sound generated in the missing spectrum
generation means is necessary in order to generate a voice resembling the voice of the user while
having the system learn in advance for speech synthesis. It is also possible to have a missing
spectrum.
[0012]
(2) Further, a second means for solving the above problems is a voice pickup system for
detecting bone conduction sound or airway sound by a pickup, wherein the sound recognition
means capable of recognizing a sound detected by the pickup; And an artificial sound generation
unit configured to generate a sound corresponding to the data of the sound recognized by the
recognition unit as an artificial sound.
10-04-2019
3
[0013]
In such a voice pickup system, as artificial sound generated by the artificial sound generation
means, it is also possible to make the system learn in advance for speech synthesis and generate
a voice resembling the voice of the user. It is also possible to generate general voices of
[0014]
In the voice pickup system as a first means for solving the problem, the bone conduction sound
or the airway sound is detected by the pickup, the sound detected by the pickup is recognized by
the voice recognition means, and the voice recognition means is recognized. The sound data is
used to generate a spectrum of frequency components when this recognized sound data is
detected by a normal microphone, and the spectrum generated in this way is compared with the
spectrum of the sound actually detected by the pickup. The missing frequency component is
found by comparison by means, and the sound of the frequency component missing from the
sound detected by the pickup is generated as a complementary sound by the missing spectrum
generation means from the result of the comparison by the comparing means, The detected
sound and the complementary sound generated by the missing spectrum generation means are
synthesized by the synthesis means and output.
[0015]
According to the voice pickup system as described above, by detecting the bone conduction
sound and the airway sound and recognizing the voice and complementing the missing spectrum,
a clear voice close to a normal voice without impairing the feature of the voice of the person. It
becomes possible to generate a signal.
Also, it is not affected by ambient noise.
[0016]
In a voice pickup system which is a second means for solving the problem, the bone conduction
sound or airway sound is detected by the pickup, the sound detected by the pickup is recognized
by the speech recognition means, and the sound recognized by the speech recognition means is
handled. Sound is generated as artificial sound by artificial sound generation means.
10-04-2019
4
[0017]
According to the above audio pickup system, bone conduction sound and airway sound are
detected, and a corresponding artificial voice is generated by data recognized as voice to
generate a clear voice signal close to normal voice. Will be able to
Also, it is not affected by ambient noise.
[0018]
An embodiment of the present invention will be described in detail with reference to the
drawings.
<Structure of Voice Pickup System (1)> First, the structure of a voice pickup system according to
an embodiment of the present invention will be described with reference to FIG.
[0019]
The pickup 1 is a detection means for detecting a bone conduction sound or an airway sound,
and a bone conduction microphone (bone conduction sound pick-up) is one for detecting a bone
conduction sound, and an airway sound microphone is one for detecting an airway sound. Do.
[0020]
The speech recognition circuit 2 is a circuit that recognizes known speech or syllables, and may
learn individual features of the user or may not perform such learning.
[0021]
Complementary sound generation circuit 3 is a kind of generation of the sound of the frequency
component missing from the sound detected by pickup 1 as complementary sound using data of
sound (speech, syllable) recognized by speech recognition circuit 2 Artificial sound generation
circuit.
10-04-2019
5
Further, the complementary sound generation circuit 3 is configured of a spectrum generation
circuit 3a, a comparison circuit 3b, and a missing spectrum generation circuit 3c.
[0022]
The spectrum generation circuit 3a uses the data of the sound recognized by the speech
recognition circuit 2 to generate a spectrum when the recognized sound is detected by a normal
microphone.
The comparison circuit 3b compares the spectrum generated by the spectrum generation circuit
3a with the spectrum of the sound detected by the pickup.
The missing spectrum generation circuit 3c generates the spectrum (lost spectrum) of the
portion corresponding to the difference of the spectrum as the complementary sound according
to the comparison result of the comparison circuit 3b.
[0023]
The synthesis circuit 4 is an output means for synthesizing and outputting the sound (bone
conduction sound, airway sound) detected by the pickup 1 and the complementary sound
generated by the complementary sound generation circuit 3.
<Operation of Voice Pickup System (1)> The operation of the voice pickup system according to
an embodiment of the present invention is roughly divided into the steps of,.
This step will be described in order.
[0024]
Detection of sound (bone conduction sound, airway sound): Using the pickup 1, bone conduction
10-04-2019
6
sound or airway sound is detected.
Recognition of sound (bone conduction sound, airway sound): The sound detected by the pickup
1 is recognized by the speech recognition circuit 2. In this case, depending on the recognition
method of the speech recognition circuit 2, whether the recognition by a single sound or the
recognition by a syllable differs, but may be any. Also, it may be one that learns and recognizes
the voice of the user, or one that does not perform learning that identifies the user.
[0025]
Spectrum generation of recognition sound: Data of recognition sound is used to generate a
spectrum (hereinafter referred to as a standard sound spectrum) which can be obtained when the
spectrum generation circuit 3a detects it with a normal microphone. For this purpose, the
spectrum generation circuit 3a has a spectrum according to the recognized sound (monophone,
syllable), and the corresponding spectrum is called according to the recognized sound.
[0026]
The spectrum in this case may have a spectrum for each user or may have a standard spectrum.
Also, in the case of having a standard spectrum, it may have several spectra such as adult males,
adult females, and children.
[0027]
Spectrum comparison between standard sound and detected sound: A comparison circuit 3b
compares the spectrum of the standard sound with the spectrum of the detected sound.
[0028]
For example, to explain with reference to FIG. 5, the spectrum A of the standard sound and the
spectrum B of the detected sound are compared for each of the recognized detected sounds to
calculate a missing spectrum C (= A−B) of the detected sound. .
[0029]
10-04-2019
7
Complementary sound (missing spectrum) generation: A missing spectrum generating circuit 3c
generates a signal of a missing spectrum according to the data of the missing spectrum obtained
by the comparison circuit 3b.
Also in this case, as with the spectrum generation circuit 3a, the generated missing spectrum
may have a spectrum for each user or may have a standard spectrum.
Also, in the case of having a standard spectrum, it may have several spectra such as adult males,
adult females, and children.
[0030]
Synthesis of Detection Sound and Complementary Sound: The synthesis circuit 4 synthesizes the
missing spectrum and the spectrum of the detection sound. By this synthesis processing, the
missing spectrum of the detected sound is added as a complementary sound, and a synthesized
sound of a spectrum equivalent to the standard spectrum is obtained. Therefore, an audio signal
equivalent to that collected by a standard microphone can be obtained.
[0031]
Note that if the complementary sound is the user's own, the synthesized sound obtained as a
result of synthesis is also the user's own, but even if the complementary sound is based on
standard data, the portion of the complementary sound is high There is very little discomfort
because it is only part of the area.
[0032]
When it is desired to generate a missing spectrum according to the voice of the user, as shown in
FIG. 2, the personal data memory 3d is provided, and the voice of the user is recorded (sampled)
in advance, It is possible to store data.
[0033]
It is also possible to determine the quality of the user's voice and automatically select a near
10-04-2019
8
standard from among a plurality of standard spectra provided for missing spectrum generation.
<Effects Obtained by Configuration (1)> According to the above-described voice pickup system,
voice detection of bone conduction sound and airway sound is performed and voice recognition
is performed to compensate for the missing spectrum, thereby impairing the feature of the voice
of the person. It becomes possible to generate clear speech signals close to ordinary speech
without.
Also, it is not affected by ambient noise.
[0034]
In addition, no complementary sound is generated for the meaningless sound of something
detected by the pickup, such as a tooth biting sound, because the meaning is not made in the
process of speech recognition. Therefore, only the low frequency component detected by the
pickup is output, so the adverse effect is small.
[0035]
<Configuration of Voice Pickup System (2)> First, the configuration of a voice pickup system
according to a second embodiment of the present invention will be described with reference to
FIG.
[0036]
The pickup 1 is a detection means for detecting a bone conduction sound or an airway sound,
and a bone conduction microphone is suitable for detecting a bone conduction sound, and an
airway sound microphone is suitable for detecting an airway sound.
[0037]
The speech recognition circuit 2 is a circuit that recognizes known speech or syllables, and may
learn individual features of the user or may not perform such learning.
[0038]
The artificial sound generation circuit 5 generates artificial sound corresponding to the sound
10-04-2019
9
detected by the pickup 1 using data of the sound (speech, syllable) recognized by the speech
recognition circuit 2.
[0039]
<Operation of Voice Pickup System (2)> The operation of the voice pickup system according to
an embodiment of the present invention is roughly divided into the following steps.
This step will be described in order.
[0040]
Detection of sound (bone conduction sound, airway sound): Using the pickup 1, bone conduction
sound or airway sound is detected.
Recognition of sound (bone conduction sound, airway sound): The sound detected by the pickup
1 is recognized by the speech recognition circuit 2.
In this case, depending on the recognition method of the speech recognition circuit 2, whether
the recognition by a single sound or the recognition by a syllable differs, but may be any.
Also, it may be one that learns and recognizes the voice of the user, or one that does not perform
learning that identifies the user.
[0041]
Spectrum generation of recognition sound: The data of recognition sound is used to generate a
standard spectrum that would be obtained when the artificial sound generation circuit 5 detects
it with a normal microphone. For this purpose, the artificial sound generation circuit 5 has a
spectrum according to the recognition sound (monophone, syllable), and the corresponding
spectrum is called according to the recognition sound.
10-04-2019
10
[0042]
The spectrum in this case may have a spectrum for each user or may have a standard spectrum.
Also, in the case of having a standard spectrum, it may be configured to switch and use several
spectra such as adult males, adult females, children, etc. Therefore, an audio signal equivalent to
that collected by a standard microphone can be obtained.
[0043]
When it is desired to generate a missing spectrum according to the user's voice, as shown in FIG.
4, the personal data memory 6 is provided and the voice of the user is recorded (sampled) in
advance to It is possible to store data.
[0044]
It is also possible to determine the quality of the user's voice and automatically select a near
standard from among a plurality of standard spectra provided for missing spectrum generation.
<Effects Obtained by Configuration (2)> According to the above-described voice pickup system,
bone-conducted sounds and airway sounds are detected, and a corresponding artificial voice is
generated based on the voice-recognized data to obtain normal voice. It is possible to generate
clear speech signals close to. Also, it is not affected by ambient noise.
[0045]
In addition, artificial sounds are not generated for meaningless sounds of those detected by the
pickup, such as tooth biting sounds, because they do not make sense in the process of speech
recognition. Therefore, there is little adverse effect because it is not output.
[0046]
10-04-2019
11
<Other Preferred Examples> Although the above-described voice pickup system is applicable to
various applications, it can be incorporated into various systems for transmitting voice in a noisy
environment. For example, by incorporating it into a device such as a mobile phone, clear
transmission becomes possible. And it also has the advantage that the other party can not hear
surrounding sounds.
[0047]
According to the voice pickup system as described above, normal speech is detected without
impairing the feature of the voice of the person by detecting the bone conduction sound and
airway sound and then performing voice recognition to complement the missing spectrum. It is
possible to generate clear speech signals close to. Also, it is not affected by ambient noise.
Therefore, it is possible to realize a voice pickup system capable of transmitting voice in a high
clarity state without picking up ambient noise.
[0048]
In addition, by detecting bone-conducted sounds and airway sounds and generating
corresponding artificial voices from the voice-recognized data, it becomes possible to generate
clear voice signals close to normal voices. Also, it is not affected by ambient noise. Therefore, it is
possible to realize a voice pickup system capable of transmitting voice in a high clarity state
without picking up ambient noise.
[0049]
Brief description of the drawings
[0050]
1 is a block diagram showing the configuration of an audio pickup system according to an
embodiment of the present invention.
[0051]
<Figure 2> It is the block diagram which shows the constitution of the deformation example of
the voice pickup system of one example of this invention.
10-04-2019
12
[0052]
3 is a block diagram showing the configuration of a voice pickup system according to a second
embodiment of the present invention.
[0053]
4 is a block diagram showing the configuration of a modification of the voice pickup system of
the second embodiment of the present invention.
[0054]
5 is a characteristic diagram schematically showing the spectrum of the voice.
[0055]
6 is a characteristic diagram showing the measurement result of the spectrum of the voice.
[0056]
Explanation of sign
[0057]
Reference Signs List 1 pickup 2 speech recognition circuit 3 complementary sound generation
circuit 3a spectrum generation circuit 3b comparison circuit 3c missing spectrum generation
circuit 4 synthesis circuit
10-04-2019
13
Документ
Категория
Без категории
Просмотров
0
Размер файла
22 Кб
Теги
description, jph08275279
1/--страниц
Пожаловаться на содержимое документа