CN111261181A - Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus - Google Patents
Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus Download PDFInfo
- Publication number
- CN111261181A CN111261181A CN202010043987.0A CN202010043987A CN111261181A CN 111261181 A CN111261181 A CN 111261181A CN 202010043987 A CN202010043987 A CN 202010043987A CN 111261181 A CN111261181 A CN 111261181A
- Authority
- CN
- China
- Prior art keywords
- sound source
- sound
- conduction microphone
- bone conduction
- processor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000004891 communication Methods 0.000 title claims abstract description 14
- 210000000988 bone and bone Anatomy 0.000 claims abstract description 97
- 230000008569 process Effects 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000009467 reduction Effects 0.000 description 24
- 230000000694 effects Effects 0.000 description 19
- 238000012360 testing method Methods 0.000 description 7
- 230000007613 environmental effect Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 3
- 230000004069 differentiation Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/02—Constructional features of telephone sets
- H04M1/03—Constructional features of telephone transmitters or receivers, e.g. telephone hand-sets
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/02—Constructional features of telephone sets
- H04M1/19—Arrangements of transmitters, receivers, or complete sets to prevent eavesdropping, to attenuate local noise or to prevent undesired transmission; Mouthpieces or receivers specially adapted therefor
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
- G10L2021/02165—Two microphones, one receiving mainly the noise signal and the other one mainly the speech signal
Abstract
The invention belongs to the technical field of acoustoelectric technology, and particularly relates to a voice recognition method, a noise recognition method, a sound pickup device and a communication device, which comprise a step S100: the bone conduction microphone collects the speaking voice of the user; s200: an air conduction microphone collects ambient sound; s300: the first sound source and the second sound source are converted into electric signals and then transmitted to the processor; s400: the processor identifies an electrical signal of a second sound source from the electrical signal of the first sound source; the bone conduction microphone and the air conduction microphone are matched to collect sound, the collected sound is transmitted to the processor to be processed, and the required sound is identified.
Description
Technical Field
The invention belongs to the technical field of acoustoelectric technology, and particularly relates to a voice recognition method, a noise recognition method, a sound pickup device and a communication device.
Background
At present, most of communication equipment such as mobile phones and earphones adopt air conduction microphones, and sound frequency is collected through sound propagation in the air, so that the functions of recording and communication are achieved. In the earphone and the mobile phone in the market, a double-microphone noise reduction scheme is adopted, wherein one microphone collects the voice of a person speaking, the other microphone collects the voice of the surrounding environment, and the voice of the surrounding environment is collected to be subjected to noise reduction algorithm processing, so that the surrounding noise and the noisy environment are reduced, and the conversation is clearer; however, the noise reduction scheme of the double microphones still has the problems of large noise of the ambient background sound and poor voice recognition effect during the call.
Disclosure of Invention
The invention aims to provide a voice recognition method, a noise recognition method, a sound pickup device and a communication device, and aims to solve the technical problem that the sound recognition effect of the device in the prior art is poor.
In order to achieve the above object, an embodiment of the present invention provides a speech recognition method, including the following steps:
s100: a first sound source collection, wherein a bone conduction microphone collects the speaking sound of a user;
s200: a second sound source collection, an air conduction microphone collects ambient sound;
s300: the sound source transmission device comprises a sound source transmission device, a first sound source and a second sound source, wherein the first sound source and the second sound source are converted into electric signals and then transmitted to a processor;
s400: sound source identification, the processor identifying the electrical signal of the second sound source from the electrical signal of the first sound source.
Optionally, after the step 400, a step S500 is further included: and sound source identification processing, wherein based on the electric signal of the first sound source, the processor identifies the electric signal corresponding to the first sound source in the second sound source and shields the electric signal corresponding to the first sound source.
Optionally, in step S300, a mode one and a mode two are further included;
in the first mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to stop working or be in a standby state;
and in the second mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to work.
Optionally, in step S400, the recognized electrical signal of the second sound source is combined with the electrical signal of the first sound source to realize high-precision speech recognition.
One or more technical solutions in the speech recognition method provided by the embodiment of the present invention have at least one of the following technical effects: the bone conduction microphone and the air conduction microphone are matched to collect sound, the collected sound is transmitted to the processor to be processed, and the required sound is identified.
The embodiment of the invention provides a noise identification method, which comprises the following steps:
s100: collecting vibration sound of equipment by a bone conduction microphone;
s200: a second sound source pick-up, an air conduction microphone, picks up sound around the device;
s300: the sound source transmission device comprises a sound source transmission device, a first sound source and a second sound source, wherein the first sound source and the second sound source are converted into electric signals and then transmitted to a processor;
s400: and identifying the sound source, wherein the processor identifies the electric signal of the first sound source according to the electric signal of the second sound source.
Optionally, after the step 400, a step S500 is further included: and sound source identification processing is carried out, wherein the processor identifies the electric signal corresponding to the first sound source in the second sound source on the basis of the electric signal of the second sound source, and shields the electric signal outside the second sound source, and the processor processes the electric signal of the second sound source and the electric signal of the first sound source to obtain a noise source.
One or more technical solutions in the noise identification method provided by the embodiment of the present invention have at least one of the following technical effects: when the device is an automobile, the bone conduction microphone collects vibration sounds of the automobile, including noise generated by bumping during the driving process of the automobile; the air conduction microphone collects ambient sound in the automobile during driving of the automobile, and the processor identifies noise generated by bumping in the ambient sound in the automobile according to the electric signal of the first sound source and the electric signal of the second sound source.
The embodiment of the invention provides a sound pickup device, which comprises a bone conduction microphone, an air conduction microphone and a processor; the bone conduction microphone and the air conduction microphone are both electrically connected with the processor so that the processor processes sound electric signals collected by the bone conduction microphone and the air conduction microphone.
Optionally, the sound pickup apparatus further comprises a housing; an accommodating space is arranged in the shell, and the bone conduction microphone, the air conduction microphone and the processor are all arranged in the accommodating space; the bone conduction earphone is characterized in that a binding surface used for being bound with skin is arranged on the outer side wall of the shell, a mounting surface opposite to the binding surface is arranged on the inner side wall of the shell, and the bone conduction earphone is connected to the mounting surface.
Optionally, the bone conduction microphone and the air conduction microphone are provided in plurality, and each of the bone conduction microphone and the air conduction microphone is electrically connected to the processor.
One or more technical solutions in the communication device provided by the embodiment of the present invention at least have one of the following technical effects: through the combined use of the bone conduction microphone and the air conduction microphone, the processor processes the collected electric signals, the sound collected by the bone conduction microphone is clear, the sound collected by the air conduction microphone is small in distortion, the characteristics of the bone conduction microphone and the air conduction microphone are complementary, and the high-quality conversation noise reduction and active noise reduction experience is obtained.
The embodiment of the invention also provides a piece of talking equipment, which comprises the sound pickup device.
One or more technical solutions in the communication device provided by the embodiment of the present invention at least have one of the following technical effects: comprises a bone conduction microphone, an air conduction microphone and a processor; the bone conduction microphone and the air conduction microphone are electrically connected with the processor so that the processor processes the sound collected by the bone conduction microphone and the air conduction microphone; through the combined use of the bone conduction microphone and the air conduction microphone, the collected electric signals are processed through the processor, the sound collected by the bone conduction microphone is clear, the sound collected by the air conduction microphone is small in distortion, the characteristics of the bone conduction microphone and the air conduction microphone are complementary, and the high-quality conversation noise reduction and active noise reduction experience is obtained.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a flowchart illustrating a speech recognition method according to an embodiment of the present invention.
FIG. 2 is a flowchart of a noise identification method according to a second embodiment of the present invention
Fig. 3 is a sectional view of a sound pickup apparatus according to an embodiment of the present invention.
Fig. 4 is a cross-sectional view of a headset according to an embodiment of the present invention.
Fig. 5 is a cross-sectional view of an earphone according to an embodiment of the present invention.
Fig. 6 is a front view of a mobile phone according to an embodiment of the present invention.
Fig. 7 is a cross-sectional view of a handset according to an embodiment of the invention.
Fig. 8 is an enlarged schematic view of a in fig. 7.
Wherein, in the figures, the respective reference numerals:
10-bone conduction microphone 20-air conduction microphone 30-processor
40-power supply 50-shell 51-accommodating space
52-binding surface 53-mounting surface 60-mobile phone shell
61-listening part 62-receiver 70-shell
100-sound pickup apparatus.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to fig. 1-8 are exemplary and intended to be used to illustrate embodiments of the invention, and should not be construed as limiting the invention.
In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the embodiments of the present invention and simplifying the description, but do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present invention, "a plurality" means two or more unless specifically limited otherwise.
In the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," "fixed," and the like are to be construed broadly, e.g., as being fixedly connected, detachably connected, or integrated; can be mechanically or electrically connected; either directly or indirectly through intervening media, either internally or in any other relationship. Specific meanings of the above terms in the embodiments of the present invention can be understood by those of ordinary skill in the art according to specific situations.
Example one
In an embodiment of the present invention, a speech recognition method is provided, as shown in fig. 1, including the following steps:
s100: a first sound source collection, wherein a bone conduction microphone collects the speaking sound of a user; specifically, the bone conduction microphone 10 collects only the speech sound of the user (i.e., user);
s200: a second sound source collection, an air conduction microphone collects ambient sound; specifically, the air conduction microphone 20 can collect other sounds of the surrounding environment within a range where the air conduction microphone 20 can collect, in addition to the speech sound of the user (i.e., user);
s300: the sound source transmission, the first sound source and the second sound source are converted into electric signals and then transmitted to the processor;
s400: sound source identification, wherein the processor identifies an electric signal of a second sound source according to an electric signal of a first sound source; the electrical signal in this embodiment may be an audio frequency, an audio data signal, an analog signal, etc.
In the embodiment of the present invention, the bone conduction microphone 10 and the air conduction microphone 20 cooperate to collect sounds, and the collected sounds are transmitted to the processor for processing, so as to identify the desired sounds.
In the embodiment of the present invention, as shown in fig. 1, after step 400, the sound collection method further includes step S500: and sound source identification processing, wherein the processor identifies the electric signal corresponding to the first sound source in the second sound source and shields the electric signal except the electric signal corresponding to the first sound source on the basis of the electric signal of the first sound source. In the embodiment of the present invention, the bone conduction microphone 10 is attached to the skin of the user and collects the speaking voice of the user, which is the first sound source, and the collected voice has high definition and low external noise; the air conduction microphone 20 picks up other ambient sounds as a second sound source; the sound that bone conduction microphone 10 and air conduction microphone 20 gathered is transmitted to the treater after the signal of telecommunication after converting into, the treater is handled the signal of telecommunication of second sound source according to the signal of telecommunication of first sound source as the basis, select the signal of telecommunication that second sound source and first sound source correspond, and shield all the other miscellaneous sound signals, the signal of telecommunication of first sound source and the signal of telecommunication of second sound source combine to compensate the shortcoming of user's pronunciation sound distortion that bone conduction microphone 10 gathered, the sound reduction degree after the combination is high, sound is clear, greatly reduced noise.
In the embodiment of the present invention, the sound collection method further includes a mode one and a mode two in step S300; in the first mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to stop working or be in a standby state; and in the second mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to work.
Specifically, the following cases are enumerated according to the usage scenarios in different modes:
when the sound pressure difference between the electric signal of the second sound source and the electric signal of the first sound source is within a preset value range, collecting the bone conduction microphone of the first sound source to work, collecting the air conduction microphone of the second sound source to work, and at the moment, keeping in a mode II;
when the sound pressure difference between the electric signal of the second sound source and the electric signal of the first sound source exceeds a preset value range, acquiring the bone conduction microphone of the first sound source to work, and acquiring the air conduction microphone of the second sound source to stop working or be in a standby state, wherein the first mode is adopted at the moment;
thirdly, when the types and the number of different sound sources collected by the second sound source exceed the preset value range, collecting the bone conduction microphone of the first sound source to work, and collecting the air conduction microphone of the second sound source to stop working or be in a standby state, wherein the first mode is adopted at the moment;
and fourthly, when the types and the number of the different sound sources collected by the second sound source are in a preset value range, and the sound pressure difference between the electric signal of the second sound source and the electric signal of the first sound source is in a preset value range, the second sound source is in a mode II.
In the embodiment of the invention, the sound collection method combines the electric signal of the second sound source after being recognized with the electric signal of the first sound source so as to realize high-precision speech recognition; the first sound source performs first voice recognition, the second sound source performs second voice recognition, double-layer voice recognition is achieved, the voice recognition effect is enhanced, and the definition is improved.
Wherein the processing of electrical signals in a processor is prior art and does not involve software modifications, as is well known to those skilled in the art, some of which are illustrated below:
firstly, testing parameters of the bone conduction microphone 10 and the air conduction microphone 20 by a microphone curve testing instrument, testing sounds with the same multiple frequencies by the bone conduction microphone 10 and the air conduction microphone 20 in a mute laboratory to obtain a pickup curve, comparing the picked sounds with the same frequencies, calculating rules and differences of the sound frequencies after the sounds are picked up by the bone conduction microphone 10 and the air conduction microphone 20 under the same frequencies by the instrument, repeatedly testing a plurality of different sound frequencies, repeatedly calculating the rules and the differences to finally form the rule parameters of the bone conduction microphone 10 and the air conduction microphone 20, identifying the same sound frequency by an algorithm chip or a processor according to the matching of the pickup frequency rules of the bone conduction microphone 10 and the air conduction microphone 20 in real time, collecting the speaking sound of a user by the bone conduction microphone 10, matching the recorded sound frequency with the sound frequency collected by the air conduction microphone 20 according to the rules, after the matching is successful, the sound frequencies except the matched sound frequency are shielded, and the noise reduction effect and the human voice recognition effect are achieved.
Secondly, because of the different structures and different purposes of different products, the frequency difference rule can be calculated by using the actual product to perform frequency difference test on searching the sound frequency rule points. In the scheme of the sound algorithm, the intermediate frequency band of the voice can be also locked, and the frequency curve range is narrowed, for example, the frequency range of the test sound is narrowed to 500 + 3000hz, which belongs to the voice intermediate frequency range, so that the frequency rule test and the rule point calculation are more accurate.
The bone conduction microphone 10 records sounds of the same frequency as the air conduction microphone 20. The bone conduction microphone 10 acquires 60DB at a 500hz position, 57.5DB at a 600hz position, 58DB at a 1000hz position, 59DB at a 1500hz position, and 65DB at a 2000hz position.
The air conduction microphone 20 picks up sound frequencies of 500hz position 65db, 560hz position 68db, 600hz position 68.5db, 1000hz position 69db, 1500hz position 72db, 2000hz position 79 db.
The difference between them is that the sound collected by the bone conduction microphone 10 is-5 db at 500hz position, 10db at 1000hz position, 13db at 1500hz position, and 14db at 2000hz position, compared with the sound collected by the air conduction microphone 20. Assuming that the person speaks, the bone conduction microphone 10 records sound frequency curves of 50db at the 500hz position, 55db at the 1000hz position, 58db at the 1500hz position, and 60db at the 2000hz position. After the difference rule compensation is performed on the voice curve acquired by the bone conduction microphone 10 according to the rule difference tested before, the voice curve of the user acquired by the air conduction microphone 20 is 55db at the 500hz position, 65db at the 1000hz position, 71db at the 1500hz position, 74db at the 2000hz position, and the algorithm chip or algorithm software is used for calculating the voice curve of the user acquired by the air conduction microphone 20, so that the voice of the user acquired by the air conduction microphone 20 is reserved after the matching is successful, and other voices are shielded and removed, thereby achieving the purposes of call noise reduction and human voice recognition.
Thirdly, sound source analysis is performed on the sound frequencies collected by the air conduction microphone 10 by using hardware or software algorithm, different sound sources generate different frequency curves, hundreds and thousands of different sound source curves can be produced at the same time, because the speaking of people can not produce a plurality of different sounds at the same time, therefore, in the sound source curves, the frequency curve of the speaking person, after the bone conduction microphone 10 and the air conduction microphone 20 are calculated according to the differentiation rule, the bone conduction microphone 10 is used for collecting the voice frequency of the speaking person to carry out differentiation rule compensation to obtain a voice frequency curve of the speaking person collected by the air conduction microphone 20, the voice compensation curve of the speaking person collected by the bone conduction microphone 20 is used for matching in a voice frequency curve library collected by the air conduction microphone 10, and after matching is successful, the voices of other frequency curves are removed, so that noise reduction and voice recognition are achieved.
Fourthly, the bone conduction microphone 10 collects the voice curve of the speaking voice of the person, the section of the curve is simulated to the air conduction microphone 20 by software or hardware to pick up the voice to obtain a new voice frequency curve, the difference point of the frequency curve is calculated by the comparison algorithm, the speaking voice of the user collected by the air conduction microphone 20 is identified, and the voice curve synthesis and identification are carried out to achieve the noise reduction and the voice identification.
In terms of hardware and software, the actual situation is that the product itself performs regular calculation with 0.001-1hz precision or higher and calculation speed of hundreds to ten thousand times per second under the support of software and hardware, the above scheme is one of implementation cases and is not limited by a fixed scheme, and a technical detailed scheme is not made due to the fact that the scheme is an example.
In the embodiment of the present invention, in step S100 of the sound collection method, the plurality of bone conduction microphones 10 are used to collect sound, so as to improve the sound collection effect, improve the sound collection accuracy, and meet different use conditions.
In the embodiment of the present invention, in step S200 of the sound collection method, the plurality of air conduction microphones 20 are used to collect sound, so that the sound collection effect is improved, the application range is wide, and different use conditions are met.
Example two
In the embodiment of the present invention, as shown in fig. 2, a noise identification method is provided, which includes the following steps:
s100: collecting a first sound source, wherein a bone conduction microphone collects vibration sound of equipment, and the equipment in the embodiment is an automobile;
s200: a second sound source collects, and an air conduction microphone collects sounds around the equipment, namely, sounds in the automobile;
s300: the sound source transmission, the first sound source and the second sound source are converted into electric signals and then transmitted to the processor;
s400: and sound source identification, wherein the processor identifies the electric signal of the first sound source according to the electric signal of the second sound source.
In the embodiment of the present invention, the bone conduction microphone 10 collects vibration sounds of the vehicle, including noise generated by bumping of the vehicle during driving; the air conduction microphone 20 collects environmental sounds inside and outside the automobile during driving of the automobile, and the processor processes the electric signal of the first sound source and the electric signal of the second sound source, thereby recognizing noise generated due to bumping in the environmental sounds inside and outside the automobile.
In the embodiment of the present invention, as shown in fig. 2, a noise identification method is provided, and after step 400, the method further includes step S500: the sound source identification processing is that based on the electric signals of the second sound source, the processor identifies the electric signals corresponding to the first sound source in the second sound source and shields the electric signals outside the second sound source, and the processor processes the electric signals collected by the second sound source and the first sound source to obtain a noise source; the processor processes the electric signal of the first sound source based on the electric signal of the second sound source, screens out the electric signal corresponding to the first sound source from the second sound source, and shields the rest of the noise electric signals; specifically, the first sound source body and the second sound source body are arranged at the same position, the bone conduction microphone and the air conduction microphone are combined into a noise identification module and are arranged in the automobile, and the bone conduction microphone and the air conduction microphone can also be arranged at intervals, so that the noise of an engine and the like can be conveniently processed in the driving process of the automobile; this scheme can a loudspeaker of adaptation in the application, above-mentioned scheme discernment noise and after handling, and the signal of telecommunication after handling sends to loudspeaker, and the noise cancellation in loudspeaker sound production and the car realizes the noise reduction effect.
EXAMPLE III
In an embodiment of the present invention, as shown in fig. 3, there is also provided a sound pickup apparatus including a bone conduction microphone 10, an air conduction microphone 20, and a processor 30; the bone conduction microphone 10 and the air conduction microphone 20 are both electrically connected with the processor 30 so that the processor 30 processes the sound electrical signals collected by the bone conduction microphone 10 and the air conduction microphone 20; through the combined use of the bone conduction microphone 10 and the air conduction microphone 20, the processor 30 processes the collected electric signals, the sound collected by the bone conduction microphone 10 is clear, the distortion of the sound collected by the air conduction microphone 20 is small, and the characteristics of the bone conduction microphone 10 and the air conduction microphone 20 are complementary, so that the excellent conversation noise reduction and active noise reduction experience is obtained. The processor 30 is well known in the art and will not be described.
In the embodiment of the present invention, as shown in fig. 3, the sound pickup apparatus further includes a housing 50; an accommodating space 51 is provided in the housing 50, and the bone conduction microphone 10, the air conduction microphone 20 and the processor 30 are all provided in the accommodating space 51; an attaching surface 52 for attaching to skin is arranged on the outer side wall of the shell 50, an installing surface 53 arranged opposite to the attaching surface 52 is arranged on the inner side wall of the shell 50, and the bone conduction headset 10 is connected to the installing surface 53; the fitting surface 52 is fitted to the skin of a human body, the bone conduction microphone 10 collects the frequency of sound through bones and transmits the collected sound to the processor 30 for processing, and the sound collection effect of the bone conduction microphone 10 is improved by the fitting surface 52.
In the embodiment of the invention, the bone conduction microphones 10 of the sound pickup device are provided in plurality, and the bone conduction microphones 10 are provided, so that the sound collection effect is improved, the application range is wide, and different use conditions are met
In the embodiment of the present invention, a plurality of air conduction microphones 20 of the sound pickup apparatus are provided, and each air conduction microphone 20 is electrically connected to the processor 30; set up a plurality of air conduction microphones 20, improve the sound collection effect, application scope is wide, satisfies different in service behavior.
In another embodiment of the present invention, as shown in fig. 2, the sound pickup apparatus 100 further includes a power source 40; the power supply 40 is electrically connected to the bone conduction microphone 10, the air conduction microphone 20, and the processor 30, respectively.
Example four
In another embodiment of the present invention, there is also provided a communication device, including the sound pickup apparatus described above; the sound pickup device can be a headset or a mobile phone.
Specifically, as shown in fig. 4, the earphone includes a housing 70, the sound pickup device 100 is disposed in the housing 70, after wearing the earphone, the housing 70 is attached to the skin, the frequency generated by the bone is transmitted to the sound pickup device 100 through the housing 70, the bone conduction microphone 10 is responsible for collecting the frequency of the speech of the human, the air conduction microphone 20 collects the external sound and the frequency of the speech of the human, the bone conduction microphone 10 collects the frequency of the speech of the human, the air conduction microphone 20 collects the frequency of the speech of the human and the frequency of the external sound, the processing by the processor 30 is based on the frequency of the speech of the human collected by the bone conduction microphone 10, the frequency of the speech collected by the air conduction microphone 20 and the frequency of the speech of the human collected by the bone conduction microphone 10 are synthesized into the speech of the user with high degree of restitution, so as to achieve the effect of reducing noise of good communication quality, when the mobile phone or the earphone is used for calling or recording in the mode, the mobile phone or the earphone can not be interfered by surrounding environment sounds, can record clearly, can also collect sound in a directional mode, and can record the human voice frequency according to the voice frequency of human speaking collected by bone conduction.
Specifically, as shown in fig. 5, the earphone is different from the above-described earphone in that a bendable holder is provided on the earphone, and an air conduction microphone 20 is provided on the holder for collecting sounds around the user.
Compared with the defects of low noise reduction performance under the noisy environment condition, low voice recognition rate of speaking of people, obvious environment noise and the like in the noise reduction of the traditional air microphone, the use of the bone conduction microphone 10 and the air conduction microphone 20 has better communication noise reduction effect under the technical scheme, is not influenced by the environment noise and increases the voice recognition rate.
Specifically, as shown in fig. 6 to 8, the mobile phone includes a mobile phone shell 60, a listening portion 61 is disposed on an outer side wall of the mobile phone shell 60, a receiver 62 is disposed on the listening portion 61, and a sound pickup device 100 is connected to the listening portion 61 and located at one side of the receiver 62; the bone conduction microphone 10 is arranged at a position close to the mobile phone receiver 62 and attached to the mobile phone shell 60, the air conduction microphone 20 can be arranged at a side face of the mobile phone shell 60 or a back face of the mobile phone shell 60, when the bone conduction microphone 10 is started to communicate with the air conduction microphone 20 for noise reduction, when a person holds the mobile phone for communication and the receiver 62 is attached to an ear, the vibration frequency of bones and muscles of the person is transmitted to the bone conduction microphone 10 through the receiver 62, the speaking sound frequency of the person is identified from the sound frequency collected by the air conduction microphone 20 according to the speaking sound frequency collected by the bone conduction microphone 10, the speaking sound frequency of the person is identified, and the environmental noise is removed through the processing of the processor 30, so that the clear and highly-restored voice of the person is obtained, and the noise reduction effect is achieved.
Compared with the traditional ANC noise reduction earphone, the technical scheme adopts the bone conduction microphone 10 to reduce the structural design difficulty and reduce the interference between the microphone and the microphone, the air conduction microphone 20 is used for collecting environmental sounds, the bone conduction microphone 10 is used for collecting voice of a conversation person, and the bone conduction microphone 10 is only used for recording the sound through the vibration frequency of an object and is not interfered by noise of the external environment, so that when the active noise reduction work is carried out, the external environmental sounds cannot be heard by the other party during the conversation simultaneously, the conversation noise reduction effect is achieved, and the voice recognition rate is increased. The noise reduction function of the processor 30 is prior art and will not be described in detail.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A speech recognition method, comprising the steps of:
s100: a first sound source collection, wherein a bone conduction microphone collects the speaking sound of a user;
s200: a second sound source collection, an air conduction microphone collects ambient sound;
s300: the sound source transmission device comprises a sound source transmission device, a first sound source and a second sound source, wherein the first sound source and the second sound source are converted into electric signals and then transmitted to a processor;
s400: sound source identification, the processor identifying the electrical signal of the second sound source from the electrical signal of the first sound source.
2. The speech recognition method according to claim 1, further comprising, after the step 400, a step S500 of: and sound source identification processing, wherein based on the electric signal of the first sound source, the processor identifies the electric signal corresponding to the first sound source in the second sound source and shields the electric signal corresponding to the first sound source.
3. The speech recognition method according to claim 1, further comprising a mode one and a mode two in step S300;
in the first mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to stop working or be in a standby state;
and in the second mode, the bone conduction microphone of the first sound source is collected to work, and the air conduction microphone of the second sound source is collected to work.
4. The speech recognition method according to claim 1, wherein in step S400, the recognized electrical signal of the second sound source is processed in combination with the electrical signal of the first sound source so that high-precision speech recognition is achieved.
5. A method of noise identification, comprising the steps of:
s100: collecting vibration sound of equipment by a bone conduction microphone;
s200: a second sound source pick-up, an air conduction microphone, picks up sound around the device;
s300: the sound source transmission device comprises a sound source transmission device, a first sound source and a second sound source, wherein the first sound source and the second sound source are converted into electric signals and then transmitted to a processor;
s400: and identifying the sound source, wherein the processor identifies the electric signal of the first sound source according to the electric signal of the second sound source.
6. The noise identification method according to claim 5, wherein after the step 400, the method further comprises a step S500: and sound source identification processing is carried out, wherein the processor identifies the electric signal corresponding to the first sound source in the second sound source on the basis of the electric signal of the second sound source, and shields the electric signal outside the second sound source, and the processor processes the electric signal of the second sound source and the electric signal of the first sound source to obtain a noise source.
7. A sound pickup apparatus comprising a bone conduction microphone, an air conduction microphone, and a processor; the bone conduction microphone and the air conduction microphone are both electrically connected with the processor so that the processor processes sound electric signals collected by the bone conduction microphone and the air conduction microphone.
8. The sound pickup apparatus of claim 7, further comprising a housing; an accommodating space is arranged in the shell, and the bone conduction microphone, the air conduction microphone and the processor are all arranged in the accommodating space; the bone conduction earphone is characterized in that a binding surface used for being bound with skin is arranged on the outer side wall of the shell, a mounting surface opposite to the binding surface is arranged on the inner side wall of the shell, and the bone conduction earphone is connected to the mounting surface.
9. The sound pickup device as claimed in any one of claims 7 to 8, wherein a plurality of the bone conduction microphones and the air conduction microphones are provided, and each of the bone conduction microphones and the air conduction microphone is electrically connected to the processor.
10. A speech communication apparatus comprising the sound pickup device according to any one of claims 7 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010043987.0A CN111261181A (en) | 2020-01-15 | 2020-01-15 | Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010043987.0A CN111261181A (en) | 2020-01-15 | 2020-01-15 | Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111261181A true CN111261181A (en) | 2020-06-09 |
Family
ID=70948961
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010043987.0A Pending CN111261181A (en) | 2020-01-15 | 2020-01-15 | Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111261181A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111935573A (en) * | 2020-08-11 | 2020-11-13 | Oppo广东移动通信有限公司 | Audio enhancement method and device, storage medium and wearable device |
| CN113113050A (en) * | 2021-05-10 | 2021-07-13 | 紫光展锐(重庆)科技有限公司 | Voice activity detection method, electronic equipment and device |
| WO2024031302A1 (en) * | 2022-08-09 | 2024-02-15 | 华为技术有限公司 | Headset, head-mounted device, and headset signal processing method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000261530A (en) * | 1999-03-10 | 2000-09-22 | Nippon Telegr & Teleph Corp <Ntt> | Speech unit |
| CN106686494A (en) * | 2016-12-27 | 2017-05-17 | 广东小天才科技有限公司 | Voice input control method of wearable equipment and the wearable equipment |
| CN108156291A (en) * | 2017-12-29 | 2018-06-12 | 广东欧珀移动通信有限公司 | Speech signal collection method, apparatus, electronic equipment and readable storage medium storing program for executing |
| CN110010143A (en) * | 2019-04-19 | 2019-07-12 | 出门问问信息科技有限公司 | A kind of voice signals enhancement system, method and storage medium |
| CN110198501A (en) * | 2018-02-27 | 2019-09-03 | 奥维通信股份有限公司 | A kind of multifunctional headphone system |
-
2020
- 2020-01-15 CN CN202010043987.0A patent/CN111261181A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000261530A (en) * | 1999-03-10 | 2000-09-22 | Nippon Telegr & Teleph Corp <Ntt> | Speech unit |
| CN106686494A (en) * | 2016-12-27 | 2017-05-17 | 广东小天才科技有限公司 | Voice input control method of wearable equipment and the wearable equipment |
| CN108156291A (en) * | 2017-12-29 | 2018-06-12 | 广东欧珀移动通信有限公司 | Speech signal collection method, apparatus, electronic equipment and readable storage medium storing program for executing |
| CN110198501A (en) * | 2018-02-27 | 2019-09-03 | 奥维通信股份有限公司 | A kind of multifunctional headphone system |
| CN110010143A (en) * | 2019-04-19 | 2019-07-12 | 出门问问信息科技有限公司 | A kind of voice signals enhancement system, method and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111935573A (en) * | 2020-08-11 | 2020-11-13 | Oppo广东移动通信有限公司 | Audio enhancement method and device, storage medium and wearable device |
| CN113113050A (en) * | 2021-05-10 | 2021-07-13 | 紫光展锐(重庆)科技有限公司 | Voice activity detection method, electronic equipment and device |
| WO2024031302A1 (en) * | 2022-08-09 | 2024-02-15 | 华为技术有限公司 | Headset, head-mounted device, and headset signal processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6408081B1 (en) | Bone conduction voice transmission apparatus and system | |
| CN111261181A (en) | Speech recognition method, noise recognition method, sound pickup device, and telephone communication apparatus | |
| EP2882203A1 (en) | Hearing aid device for hands free communication | |
| US10277750B2 (en) | Method and system for improving echo in hands-free call of mobile terminal | |
| US11510019B2 (en) | Hearing aid system for estimating acoustic transfer functions | |
| EP3883266A1 (en) | A hearing device adapted to provide an estimate of a user's own voice | |
| US20210352417A1 (en) | Hearing device configured to utilize non-audio information to process audio signals | |
| US20190104370A1 (en) | Hearing assistance device | |
| CN111683319A (en) | Call pickup noise reduction method, earphone and storage medium | |
| CN111565354B (en) | Testing method and testing system for bone conduction earphone | |
| CN114339569B (en) | Method and system for obtaining vibration transfer function | |
| EP2482566B1 (en) | Method for generating an audio signal | |
| US20220103952A1 (en) | Hearing aid comprising a record and replay function | |
| CN112367599B (en) | Hearing aid system with cloud background support | |
| CN116801157A (en) | Wireless earphone assembly and signal processing method thereof | |
| US20230028004A1 (en) | Systems and methods for obtaining vibration transfer functions | |
| US20220174428A1 (en) | Hearing aid system comprising a database of acoustic transfer functions | |
| CN115314823A (en) | Hearing aid method, system and equipment based on digital sounding chip | |
| US20240114296A1 (en) | Hearing aid comprising a speaker unit | |
| CN219843713U (en) | Bone conduction earphone with microphone for reducing noise in conversation | |
| US20230388721A1 (en) | Hearing aid system comprising a sound source localization estimator | |
| US20230054213A1 (en) | Hearing system comprising a database of acoustic transfer functions | |
| US20230421971A1 (en) | Hearing aid comprising an active occlusion cancellation system | |
| EP4210348A1 (en) | A method for monitoring and detecting if hearing instruments are correctly mounted | |
| EP4199541A1 (en) | A hearing device comprising a low complexity beamformer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |
|
| RJ01 | Rejection of invention patent application after publication |