Microphone unit

By designing a microphone unit with local recognition capabilities and utilizing wireless communication and frequency characteristic correction, the problem of identifying the speaker in the absence of a network was solved, achieving safe and accurate speaker identification and power consumption optimization.

CN114080641BActive Publication Date: 2024-11-01HOSIDEN CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202080051540.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-07-17
Filing Date
2020-06-09
Publication Date
2024-11-01
Estimated Expiration
2040-06-09

AI Technical Summary

Technical Problem

Existing technologies require communication infrastructure to identify the speaker of a voice and have failed to effectively identify the speaker.

Method used

Design a microphone unit with sound data acquisition, registration, evaluation, and comparison functions. It can perform local identification through wireless communication, securely identify the speaker without a server, and improve the comparison rate through frequency characteristic correction and parameter optimization.

Benefits of technology

It enables secure identification of the speaker even without a network connection, reduces power consumption, increases the matching rate, and can automatically adjust parameters to improve recognition accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114080641B_ABST
    Figure CN114080641B_ABST
Patent Text Reader

Abstract

The microphone unit (1) comprises: a sound data acquisition unit (11) for acquiring sound as sound data; a sound data registration unit (12) for registering comparison sound data with feature points extracted from the sound data; an evaluation sound data acquisition unit (13) for acquiring sound input to the first microphone (10) as evaluation sound data; a comparison unit (14) for comparing whether a speaker of the sound based on the evaluation sound data is the speaker of the sound based on the comparison sound data based on the comparison sound data and the feature points extracted from the evaluation sound data; and a comparison result output unit (15) for outputting the comparison result of the comparison unit (14).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a microphone unit capable of determining whether a sound input to a first microphone is the voice of an expected speaker. Background Art

[0002] Conventionally, for example, a technique for providing instructions to an electric appliance using voice is used. In such a technique, voice recognition is performed. As a technique related to voice recognition, for example, there is a technique described in Patent Document 1, which shows the source below.

[0003] Patent document 1 discloses a voice dialogue system. The voice dialogue system is composed of a voice dialogue device and a voice recognition server. The voice dialogue device recognizes the voice input to the voice input unit and sends the voice input to the voice recognition server. The voice recognition server recognizes the voice received from the voice dialogue device. The voice dialogue device is configured to output a response based on the recognition result of the voice dialogue device and then output a response based on the difference between the recognition result of the voice dialogue device and the recognition result of the voice recognition server.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent document 1: Japanese Patent Application Publication No. 2018-45190. Summary of the invention

[0007] Problems to be solved by the invention

[0008] In the technology described in Patent Document 1, the voice recognition process is performed not only in the voice dialogue device but also in the voice recognition server. Therefore, communication is required when performing voice recognition, and it cannot be used in a situation where there is no communication infrastructure. In addition, the technology described in Patent Document 1 is a technology for performing voice recognition, and it is not envisioned to recognize the speaker of the voice.

[0009] Therefore, a microphone unit is required that can recognize the speaker of a voice without requiring a network in the recognition process.

[0010] Solutions to Solve Problems

[0011] The characteristic structure of the microphone unit of the present invention lies in the following aspects: a microphone unit capable of determining whether a sound input to a first microphone is the sound of an expected speaker, the microphone unit comprising: a sound data acquisition unit for acquiring sound as sound data; a sound data registration unit for registering comparison sound data with feature points extracted from the sound data; an evaluation sound data acquisition unit for acquiring sound input to the first microphone as evaluation sound data; a comparison unit for comparing whether the speaker of the sound based on the evaluation sound data is the speaker of the sound based on the comparison sound data based on the comparison sound data and the feature points extracted from the evaluation sound data; and a comparison result output unit for outputting the comparison result of the comparison unit, the comparison sound data being produced by a device different from the device equipped with the first microphone, and the device equipped with the first microphone and the different device exchanging the comparison sound data using wireless communication.

[0012] If such a characteristic structure is made, it is possible to perform the comparison without having a comparison server. That is, since it is a so-called local recognition process, the comparison can be performed safely in terms of security. In addition, the speaker of the pre-registered voice can be easily specified.

[0013] Furthermore, it is preferable that, when the comparison unit is in the sleep state, the sleep state is ended using the acquisition of the evaluation audio data by the evaluation audio data acquisition unit as a trigger.

[0014] If such a structure is adopted, power consumption can be reduced.

[0015] In addition, preferably, the sound data acquired by the sound data acquisition unit is sound input to a second microphone provided in a device different from the device equipped with the first microphone, and further comprises: an evaluation unit for evaluating the frequency characteristics of the first microphone and the frequency characteristics of the second microphone before inputting sound to both the first microphone and the second microphone; and a correction unit for correcting the frequency characteristics of one of the first microphone and the second microphone in such a manner that the frequency characteristics of the other are consistent with the frequency characteristics of the other.

[0016] With such a configuration, by making the frequency characteristics of the first microphone and the second microphone the same, it is possible to improve the comparison rate when the same user inputs voice using the first microphone and the second microphone.

[0017] In addition, preferably, the present invention further comprises: a test sound data acquisition unit, which acquires the sound of a speaker who emits the sound of the control sound data through the first microphone as the test sound data after registering the control sound data and before acquiring the evaluation sound data; and a calculation unit, which calculates a comparison rate of the speaker based on the test sound data while changing the comparison parameters used in the comparison based on the test sound data and the control sound data, and the comparison unit performs the comparison based on the comparison parameters when the highest comparison rate is obtained among the comparison rates calculated by the calculation unit.

[0018] Generally, when the false recognition rate is reduced, the comparison rate also decreases, but there are some comparison parameters that only reduce the false recognition rate but not the comparison rate. Therefore, as in this configuration, by finding the comparison parameters while calculating the comparison rate, it is possible to set appropriate comparison parameters.

[0019] In addition, it is preferable that the comparison parameter is an amplification factor for amplifying at least one of the test audio data and the comparison audio data.

[0020] With such a configuration, for example, by changing the input range of at least one of the first microphone and the second microphone, the comparison rate can be improved.

[0021] Preferably, the device further comprises a parameter changing unit that automatically changes a parameter of the first microphone based on the comparison parameter when a voice based on the evaluation voice data is input to the first microphone.

[0022] If such a structure is made, the sound level of the sound input from the first microphone can be automatically suppressed and changed to a level that increases the comparison rate. Therefore, the comparison rate can be automatically increased. In addition, it is also possible to record a certain time, for example, and automatically change the sound level based on the average sound level during the time.

[0023] Furthermore, it is preferable that the microphone unit recognizes a speaker of the sound input to the first microphone based on the comparison result of the comparison unit.

[0024] With such a configuration, it is possible to identify the speaker of the sound input by the microphone unit used by a plurality of users, thereby improving convenience.

[0025] In addition, it is preferable that the utterance content of the sound input to the first microphone is estimated, and an operation instruction is issued to a device equipped with the first microphone based on the estimated content.

[0026] With such a configuration, the operation of the device equipped with the first microphone can be controlled based on the operation command. Therefore, for example, the operation of the device equipped with the first microphone can be controlled hands-free, thereby improving convenience. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a diagram showing the structure of the microphone unit according to the first embodiment.

[0028] Figure 2 It is a diagram showing the structure of a microphone unit according to a second embodiment.

[0029] Figure 3 It is a diagram showing the structure of a microphone unit according to a third embodiment.

[0030] Figure 4 It is a diagram showing an application example of the microphone unit. DETAILED DESCRIPTION

[0031] 1. First Implementation Method

[0032] The microphone unit of the present invention is configured to be able to determine whether or not an input voice is the voice of an expected speaker. Hereinafter, the microphone unit 1 of this embodiment will be described.

[0033] Figure 1 Schematically shows the structure of the microphone unit 1 of this embodiment. Figure 1 As shown, the microphone unit 1 includes a first microphone 10, a voice data acquisition unit 11, a voice data registration unit 12, an evaluation voice data acquisition unit 13, a comparison unit 14, and a comparison result output unit 15. These functional units are constructed by hardware or software or both with the CPU as a core component in order to perform the above-mentioned determination processing.

[0034] The first microphone 10 receives sound. The first microphone 10 is a microphone element, and the structure is not particularly limited. For example, at least one of an electret condenser microphone (ECM), an analog MEMS (Micro-Electro-Mechanical System) microphone, and a digital MEMS (Micro-Electro-Mechanical System) microphone may be used. The sound input to the first microphone 10 is converted into sound data as an electrical signal by the first microphone 10.

[0035] The sound data acquisition unit 11 acquires sound as sound data. In the present embodiment, the sound data acquired by the sound data acquisition unit 11 is sound input to the second microphone 2A, which is provided in a device 2 different from the device equipped with the first microphone 10. The device equipped with the first microphone 10 is the microphone unit 1 in the present embodiment. Therefore, the second microphone 2A is provided separately (separately) from the microphone unit 1. The second microphone 2A, like the first microphone 10, may also be configured using at least one of an electret condenser microphone (ECM), an analog MEMS (Micro-Electro-Mechanical System) microphone, a digital MEMS (Micro-Electro-Mechanical System) microphone, and the like. The sound input to the second microphone 2A is converted into sound data as an electrical signal by the second microphone 2A. The sound data acquisition unit 11 acquires the sound data generated by the conversion by the second microphone 2A.

[0036] The sound data registering unit 12 registers sound data for comparison in which feature points are extracted from the sound data generated by the second microphone 2A. The sound data generated by the second microphone 2A is data generated by converting (digitizing) the sound input to the second microphone 2A as described above. The feature point is a feature of the electrical signal (sound data), such as a period, a peak value, a half-width, etc. Therefore, the sound data for comparison is equivalent to data in which the feature of the electrical signal generated by converting the sound input to the second microphone 2A is extracted. When such sound data for comparison is the main sound data used by the microphone unit 1 to realize the function of determining whether the sound input to the first microphone 10 is the sound of the expected speaker, such sound data for comparison is recorded in the sound data registering unit 12.

[0037] The sound data for comparison is produced by a device different from the device (microphone unit 1) equipped with the first microphone 10. The device different from the device equipped with the first microphone 10 is a device different from the microphone unit 1. Specifically, it corresponds to the device 2 equipped with the second microphone 2A or the device 3 other than the microphone unit 1 and the device 2. In the present embodiment, the sound data for comparison is generated by a sound data generation unit 3A for comparison provided in the device 3. The sound data for comparison is exchanged with the device different from the device equipped with the first microphone 10 by wireless communication. The wireless communication corresponds to LAN communication such as Wi-Fi (registered trademark) or short-range wireless communication such as Bluetooth (registered trademark). Therefore, in the present embodiment, the microphone unit 1 transmits the sound data for comparison from the device 3 (the sound data generation unit 3A for comparison of the device 3) via such wireless communication. In addition, in the device 2 equipped with the second microphone 2A, when the sound data for comparison is generated, the sound data generation unit 3A for comparison can be included in the device 2.

[0038] When a device 3 other than the microphone unit 1 and the device 2 generates sound data for comparison, the sound data acquisition unit 11 can transmit the sound data to the device 3 other than the microphone unit 1 and the device 2 using wireless communication, and the sound data registration unit 12 can acquire the sound data for comparison generated in the device 3 other than the microphone unit 1 and the device 2 using wireless communication.

[0039] In addition, when the device 2 generates voice data for comparison, the voice data acquisition unit 11 can also be configured to transmit the voice data to the device 2 using wireless communication, and the voice data registration unit 12 can obtain the voice data for comparison generated in the device 2 using wireless communication. It can also be configured to include the voice data acquisition unit 11 in the device 2, and the voice data registration unit 12 can obtain the voice data for comparison generated in the device 2 based on the voice data acquired by the voice data acquisition unit 11 using wireless communication.

[0040] For example, in a device such as a well-known so-called smart speaker, the sound input to the microphone is digitized, and the digitized sound data is sent to an external device (server) via an Internet line. On the other hand, in the microphone unit 1 of the present embodiment, as described above, the comparison sound data generated by extracting feature points from the sound data by the comparison sound data generating unit 3A is sent to the paired device via wireless communication.

[0041] The evaluation sound data acquisition unit 13 acquires the sound input to the first microphone 10 as the evaluation sound data. As described above, the sound input to the first microphone 10 is converted by the first microphone 10 into sound data as an electrical signal. This sound data is equivalent to the evaluation sound data. Therefore, the evaluation sound data acquisition unit 13 acquires the evaluation sound data generated by converting the sound input to the first microphone 10 into sound data as an electrical signal by the first microphone 10.

[0042] The comparison unit 14 compares whether the speaker of the sound based on the evaluation sound data is the speaker of the sound based on the comparison sound data based on the comparison sound data, based on the comparison sound data and the feature points extracted from the evaluation sound data. The comparison sound data is registered and recorded in the sound data registration unit 12. The evaluation sound data is acquired by the evaluation sound data acquisition unit 13. The feature points extracted from the evaluation sound data are features of the evaluation sound data as an electrical signal, such as period, peak value, half-height width, etc. Such feature points can be configured to be generated by a device different from the microphone unit 1 and transmitted via wireless communication, similar to the comparison sound data. Of course, it can also be configured so that the comparison unit 14 extracts feature points. The sound based on the evaluation sound data is the sound input to the first microphone 10, that is, the sound converted by the first microphone 10 into the evaluation sound data. In the present embodiment, the sound based on the comparison sound data is the sound input to the second microphone 2A, that is, the sound converted by the second microphone 2A into the comparison sound data.

[0043] Therefore, the comparison unit 14 compares whether the speaker of the sound input to the first microphone 10, i.e., the sound converted by the first microphone 10 into the evaluation sound data, and the speaker of the sound input to the second microphone 2A, i.e., the sound converted by the second microphone 2A into the comparison sound data, are the same person, based on the comparison sound data recorded in the sound data registration unit 12 and the feature points extracted from the evaluation sound data acquired by the evaluation sound data acquisition unit 13.

[0044] For example, the characteristic parts (equivalent to the above-mentioned "characteristic points") of the evaluation voice data and the control voice data can be compared, and the consistent points and different points can be extracted, and such a comparison can be performed based on the consistency calculated according to the ratio of the consistent points and the different points. Specifically, when the consistency is greater than a preset value, it can be determined that the speaker of the voice converted into the evaluation voice data and the speaker of the voice converted into the control voice data are the same person, and when the consistency is less than a preset value, it can be determined that the speaker of the voice converted into the evaluation voice data and the speaker of the voice converted into the control voice data are not the same person. Of course, it can also be performed using a method different from such a method (for example, a known voiceprint analysis, etc.).

[0045] Here, the comparison performed by the comparison unit 14 requires computational processing, and therefore, power consumption increases when the comparison unit 14 is always set to an operating state. Therefore, it is preferred that the comparison unit 14 is set to an operating state only when the comparison is performed, and is set to a sleep state when the comparison is not performed. In the above case, when the comparison unit 14 is in a sleep state, for example, the acquisition of the evaluation sound data by the evaluation sound data acquisition unit 13 may be used as a trigger to end the sleep state. The acquisition of the evaluation sound data by the evaluation sound data acquisition unit 13 may be indicated by transmitting information indicating that the evaluation sound data has been acquired by the evaluation sound data acquisition unit 13 to the comparison unit 14, or by transmitting the evaluation sound data to the comparison unit 14. In addition, it may also be configured to transmit information indicating that the first microphone 10 has detected a sound (VoiceActive Detection: voice activity detection) to the comparison unit 14 via the evaluation sound data acquisition unit 13.

[0046] Thus, the comparison unit 14 is set to the operating state only when the comparison unit 14 performs comparison, so that the power consumption can be reduced in other states. Such a structure can be achieved by setting the operating frequency of the comparison unit 14 in sleep to be lower than the operating frequency in operation. In addition, in order to achieve such different operating frequencies, for example, the comparison unit 14 can be operated according to an external clock (external clock) from the comparison unit 14 in sleep, and can be operated according to a clock (internal clock) generated by the comparison unit 14 in operation.

[0047] Alternatively, instead of the above configuration, the comparison unit 14 may be configured to cancel the sleep state in response to a button operation (switch operation) by the user.

[0048] The comparison result output unit 15 outputs the comparison result of the comparison unit 14. The comparison result of the comparison unit 14 is a determination result of whether the speaker of the sound converted into the evaluation sound data and the speaker of the sound converted into the comparison sound data are the same person. That is, it is a determination result of whether the speaker of the sound input to the first microphone 10 and the speaker of the sound input to the second microphone 2A are the same person. The comparison result output unit 15 outputs such a determination result to a display device, and the display device displays it. Alternatively, such a determination result may be output to a speaker, and the speaker may notify it. In addition, the determination result may be output to other control devices, and the other control devices may be used for control.

[0049] For example, the speaker of the sound input to the first microphone 10 can be identified based on the comparison result of the comparison unit 14. In the above case, the speaker identification result can be output by a speaker or a display device. In addition, the identification result can be output to another control device and the other control device can be used for control.

[0050] 2. Second Implementation Method

[0051] Next, the second embodiment is described. The microphone unit 1 of the second embodiment is different from the microphone unit 1 of the first embodiment in that it includes an evaluation unit 20 and a correction unit 21. Other than this, the second embodiment is the same as the first embodiment, and therefore, the description will focus on the differences.

[0052] Figure 2 Schematically shows the structure of the microphone unit 1 of this embodiment. Figure 2 As shown, the microphone unit 1 of the present embodiment includes a first microphone 10, a voice data acquisition unit 11, a voice data registration unit 12, an evaluation voice data acquisition unit 13, a comparison unit 14, a comparison result output unit 15, an evaluation unit 20, and a correction unit 21. The evaluation unit 20 and the correction unit 21 are constructed by hardware or software or both with the CPU as the core component in order to perform the above-mentioned determination processing, similarly to the other functional units described in the first embodiment.

[0053] The evaluation unit 20 evaluates the frequency characteristics of the first microphone 10 and the frequency characteristics of the second microphone 2A before the sound is input to both the first microphone 10 and the second microphone 2A. Before the sound is input to both the first microphone 10 and the second microphone 2A means before the sound is input to both the first microphone 10 and the second microphone 2A.

[0054] That is, before sound is input to both the first microphone 10 and the second microphone 2A, it is equivalent to a state in which no sound is input to both the first microphone 10 and the second microphone 2A, a state in which sound is input only to the first microphone 10, and a state in which sound is input only to the second microphone 2A, which is a state in which no sound is input to at least one of the first microphone 10 and the second microphone 2A.

[0055] The frequency characteristics of the first microphone 10 and the frequency characteristics of the second microphone 2A are predetermined for each microphone, and therefore, such frequency characteristics may be stored in a storage unit (not shown) and acquired by the evaluation unit 20, or the evaluation unit 20 may actually power on the first microphone 10 or the second microphone 2A to perform frequency analysis and acquire the frequency characteristics. The evaluation unit 20 calculates the difference between the acquired frequency characteristics of the first microphone 10 and the acquired frequency characteristics of the second microphone 2A.

[0056] The correction unit 21 performs correction so that the frequency characteristic of one of the first microphone 10 and the second microphone 2A matches the frequency characteristic of the other. As described above, the frequency characteristic of the first microphone 10 and the frequency characteristic of the second microphone 2A are evaluated by the evaluation unit 20. In addition, the evaluation by the evaluation unit 20 is performed in a state where no sound is input to at least one of the first microphone 10 and the second microphone 2A.

[0057] When sound is not input to both the first microphone 10 and the second microphone 2A, one of the first microphone 10 and the second microphone 2A in the state where sound is not input is equivalent to one of the first microphone 10 and the second microphone 2A, and when sound is not input to one of the first microphone 10 and the second microphone 2A, one of the first microphone 10 and the second microphone 2A in the state where sound is not input is equivalent to the microphone of the first microphone 10 and the second microphone 2A to which sound is not input.

[0058] In addition, when sound is not input to both the first microphone 10 and the second microphone 2A, the other of the first microphone 10 and the second microphone 2A in the state where sound is not input is equivalent to the other of the first microphone 10 and the second microphone 2A, and when sound is not input to one of the first microphone 10 and the second microphone 2A, the other of the first microphone 10 and the second microphone 2A in the state where sound is not input is equivalent to the microphone of the first microphone 10 and the second microphone 2A to which sound is input.

[0059] Therefore, when no sound is input to both the first microphone 10 and the second microphone 2A, the correction unit 21 performs correction in a manner that makes the frequency characteristics of one of the first microphone 10 and the second microphone 2A consistent with the frequency characteristics of the other of the first microphone 10 and the second microphone 2A, and when no sound is input to one of the first microphone 10 and the second microphone 2A, the correction unit 21 performs correction in a manner that makes the frequency characteristics of the microphone of the first microphone 10 and the second microphone 2A to which no sound is input consistent with the frequency characteristics of the microphone of the first microphone 10 and the second microphone 2A to which sound is input.

[0060] With the above configuration, the frequency characteristics of at least the microphone that receives sound later and the microphone that receives sound earlier among the first microphone 10 and the second microphone 2A can be made consistent, thereby reducing errors in comparison caused by differences in the frequency characteristics of the microphones.

[0061] 3. Third Implementation Method

[0062] Next, the third embodiment will be described. The microphone unit 1 of the third embodiment is different from the microphone unit 1 of the first embodiment in that it includes a test sound data acquisition unit 30, a calculation unit 31, and a parameter change unit 40. Other than this, it is the same as the first embodiment, so the description here mainly focuses on the different aspects.

[0063] Figure 3 Schematically shows the structure of the microphone unit 1 of this embodiment. Figure 3 As shown, the microphone unit 1 of the present embodiment includes a first microphone 10, a voice data acquisition unit 11, a voice data registration unit 12, an evaluation voice data acquisition unit 13, a comparison unit 14, a comparison result output unit 15, a test voice data acquisition unit 30, a calculation unit 31, and a parameter change unit 40. The test voice data acquisition unit 30, the calculation unit 31, and the parameter change unit 40 are constructed by hardware or software or both with the CPU as the core component in order to perform the above-mentioned determination process, similarly to the other functional units described in the first embodiment.

[0064] The test voice data acquisition unit 30 acquires the voice of the speaker who uttered the voice for the control voice data from the first microphone 10 as the test voice data after registering the control voice data and before acquiring the evaluation voice data from the first microphone 10. After registering the control voice data is after the control voice data generated by converting the voice input to the second microphone 2A is registered in the voice data registration unit 12. Before acquiring the evaluation voice data is before the evaluation voice data acquisition unit 13 acquires the evaluation voice data generated by converting the voice input to the first microphone 10. The voice of the speaker who uttered the voice for the control voice data from the first microphone 10 is the same as the voice of the speaker who uttered the voice that became the basis of the control voice data registered in the voice data registration unit 12, that is, the voice acquired from the first microphone 10.

[0065] Therefore, during the period from when the control sound data generated by converting the sound input to the second microphone 2A is registered in the sound data registration unit 12 to when the evaluation sound data generated by converting the sound input to the first microphone 10 is acquired by the evaluation sound data acquisition unit 13, the test sound data acquisition unit 30 converts the sound input by the same speaker as the speaker who uttered the sound that serves as the basis of the control sound data registered in the sound data registration unit 12 into test sound data from the first microphone 10 and acquires the data.

[0066] The calculation unit 31 calculates the comparison rate of the speaker based on the test voice data while changing the comparison parameters used in the comparison based on the test voice data and the control voice data. The calculation unit 31 acquires the test voice data from the test voice data acquisition unit 30 and acquires the control voice data from the voice data registration unit 12. The parameters used in the comparison are parameters for correcting at least one of the test voice data and the control voice data so that the speaker of the voice that is the basis of the test voice data is the speaker of the voice that is the basis of the control voice data.

[0067] In this embodiment, the comparison parameter corresponds to the amplification factor of at least one of the test sound data and the comparison sound data. For example, at least one of the test sound data in the first microphone 10 and the comparison sound data in the second microphone 2A may be amplified so that the amplitudes of the test sound data and the comparison sound data are equal to each other. The amplification factor in this case corresponds to the comparison parameter.

[0068] The calculation unit 31 amplifies one or both of the input test voice data and the control voice data while sequentially changing the amplification factor, and calculates whether the speaker of the voice that is the basis of the test voice data and the speaker of the voice that is the basis of the control voice data are the same speaker. The calculation unit 31 stores such calculation results.

[0069] The comparison parameter (amplification factor) at the highest comparison rate among the comparison rates calculated by the calculation unit 31 is transmitted to the parameter changing unit 40, and the comparison parameter is set to at least one of the first microphone 10 and the second microphone 2A. Then, the comparison unit 14 performs comparison based on the test voice data and the comparison voice data to which the comparison parameter at the highest comparison rate set by the parameter changing unit 40 among the comparison rates calculated by the calculation unit 31 is applied. With such a configuration, it is possible to reduce erroneous comparison in which it is determined that the speaker of the voice based on the evaluation voice data is not the same speaker as the speaker of the voice based on the comparison voice data.

[0070] In the present embodiment, the change of the comparison parameter is configured such that the parameter changing unit 40 automatically changes the parameter of the first microphone 10 based on the comparison parameter when the sound based on the evaluation sound data is input to the first microphone 10. Thus, the comparison parameter can be easily changed, and the comparison can be performed using the evaluation sound data which can reduce the erroneous comparison in which the speaker of the sound based on the evaluation sound data is determined to be different from the speaker of the sound based on the comparison sound data.

[0071] 4. Control treatment

[0072] Next, as specific application examples of the microphone unit 1, Figure 4 The unlocking of the door shown in FIG. 1 is described as an example. First, the user 100 inputs a sound of a predetermined word into the second microphone 2A of a mobile terminal (an example of the device 2) such as a smartphone (#1). Such input of sound via the smartphone can be performed by installing an application in the smartphone in advance.

[0073] The sound input to the second microphone 2A is converted into sound data and transmitted to the microphone unit 1 via the communication function (e.g., wireless communication) of the smartphone. Such conversion into sound data is also preferably performed using an application of the smartphone. The sound data acquisition unit 11 of the microphone unit 1 acquires the sound data, and the sound data for comparison, which extracts feature points from the sound data by the sound data generation unit 3A, is registered as sound data for comparison by the sound data registration unit 12 (#2).

[0074] In this state, the user 100 inputs a voice to the first microphone 10 of the microphone unit 1 (#3). At this time, the voice can be input by uttering the above-mentioned specific word. In addition, at this point in time, the microphone unit 1 may not particularly specify whether the person inputting the voice is the user 100. The input voice is converted into evaluation voice data by the first microphone 10, and the evaluation voice data acquisition unit 13 acquires it. The comparison unit 14 compares the feature points extracted from the evaluation voice data with the comparison voice data (#5).

[0075] When the comparison unit 14 determines that the speaker of the voice based on the evaluation voice data is the speaker of the voice based on the comparison voice data, the comparison result output unit 15 outputs a signal to the locking unit indicating that the speakers of the two voices are the same person, that is, outputs a signal to the locking unit indicating that the speaker of the voice based on the evaluation voice data is the speaker of the voice based on the comparison voice data, that is, the user 100. Thus, the lock is released (#6).

[0076] On the other hand, in the state #2, a user 150 who is different from the user 100 inputs a voice to the first microphone 10 of the microphone unit 1 (#7). In addition, at this point in time, the microphone unit 1 may not particularly specify whether the person inputting the voice is the user 150. The input voice is converted into evaluation voice data by the first microphone 10, and the evaluation voice data acquisition unit 13 acquires it. The comparison unit 14 compares the evaluation voice data with the comparison voice data (#8).

[0077] When the comparison unit 14 determines that the speaker of the voice based on the evaluation voice data is not the speaker of the voice based on the comparison voice data, the comparison result output unit 15 outputs a signal to the locking unit indicating that the speakers of the two voices are not the same person, that is, outputs a signal to the locking unit indicating that the speaker of the voice based on the evaluation voice data is not the speaker of the voice based on the comparison voice data, that is, the user 100. In this case, the lock is not released, and the lock state is maintained (#9). In addition, in the above case, it is also possible to output a signal to the notification unit (not shown) indicating that the speaker of the voice based on the evaluation voice data is not the speaker of the voice based on the comparison voice data, that is, the user 100, and the notification unit outputs a sound or light indicating that the speakers are different to perform notification.

[0078] Furthermore, if the word in the input voice is different from the word in the voice of the comparison voice data, it can be determined that the user 150 who uttered the voice this time is not the user 100 .

[0079] 5. Other implementation methods

[0080] In the first embodiment, when the comparison unit 14 is in the sleep state, the acquisition of the evaluation audio data by the evaluation audio data acquisition unit 13 is used as a trigger to end the sleep state. However, the comparison unit 14 may not be in the sleep state.

[0081] In the above-mentioned second embodiment, the microphone unit 1 is described as having the functional parts of the first microphone 10, the sound data acquisition part 11, the sound data registration part 12, the evaluation sound data acquisition part 13, the comparison part 14, the comparison result output part 15, the evaluation part 20, and the correction part 21, but it can also be composed of the test sound data acquisition part 30 and the calculation part 31, and further, it can also be composed of the parameter change part 40 in addition to the test sound data acquisition part 30 and the calculation part 31.

[0082] In the third embodiment described above, the microphone unit 1 is described as including the parameter changing section 40 , but the microphone unit 1 may be configured without including the parameter changing section 40 .

[0083] In the above embodiment, an example of a case where there is only one first microphone 10 is given for explanation, but it is also possible to have a configuration with a plurality of first microphones 10. In the above case, it is preferred that only sounds from the expected direction can be input to each of the first microphones. Thus, only a specifically specified sound can be recognized, and it is easy to check the speaker.

[0084] In the second embodiment described above, the frequency characteristics of the first microphone 10 and the second microphone 2A are evaluated, and the correction unit 21 is explained so that the frequency characteristics of one of the first microphone 10 and the second microphone 2A are corrected to match the frequency characteristics of the other. Instead, first, the user 100 inputs a voice to the first microphone 10, and the microphone unit 1 transfers the voice input to the first microphone 10 to the device 2 by wireless communication. Then, the second microphone 2A of the device 2 can also be configured to acquire voice data based on the voice transferred from the microphone unit 1, and register the voice data as voice data for comparison. With such a configuration, the voice data for comparison can be generated directly using the voice input to the first microphone 10, so that correction such as making the frequency characteristics consistent can be unnecessary. In addition, even with the above configuration, it is possible to prevent erroneous comparison based on the difference in the frequency characteristics of the first microphone 10 and the second microphone 2A. In addition, it is possible to effectively use the function of the device 2 having the second microphone 2A (the function of transferring to the voice data acquisition unit 11) without giving the first microphone 10 the function of transferring to the voice data acquisition unit 11, and it is not necessary to calibrate the frequency characteristics in advance, so it is effective.

[0085] In the above embodiment, the microphone unit 1 is used for unlocking a door lock, but it can also be used for, for example, a door lock of a vehicle, or starting of a power device (such as an engine or motor) of a vehicle, or a device installed in the vehicle (a hands-free microphone, or a speaker-microphone integrated box, or an external voice recognition microphone, an internal voice recognition microphone). In addition, it is not limited to vehicles, and can also be used for smart speakers, microphones for housing assembly, surveillance cameras, intercoms, home appliances (TVs, refrigerators, rice cookers, microwave ovens, etc.), bathtub remote controls, etc. In other words, it can be said that the microphone unit 1 estimates the utterance content of the sound input to the first microphone 10 and issues an operation instruction to the device (microphone unit 1) equipped with the estimated first microphone 10.

[0086] In the above embodiment, an example in which the first microphone 10 and the second microphone 2A are different microphones has been described. However, the first microphone 10 and the second microphone 2A may be the same microphone.

[0087] In the above embodiment, in "4. Comparison Processing", it is described as "the user 100 inputs a sound of a predetermined word to the second microphone 2A of a mobile terminal such as a smartphone (an example of the device 2) (#1)", but it can be configured so that the user 100 inputs a sound of a predetermined word to the first microphone 10, and the sound input to the first microphone 10 is transmitted to the second microphone 2A via wireless communication to generate the comparison audio data. In addition, it is described as "the comparison unit 14 compares the feature points extracted from the evaluation audio data with the comparison audio data (#5)", but it can also be configured so that the feature points extracted from the evaluation audio data are also performed by the second microphone 2A. In either case, it can be configured so that each data or feature point is transmitted via wireless communication.

[0088] Industrial Applicability

[0089] The present invention can be used for a microphone unit that can determine whether the sound input to the first microphone is the voice of an expected speaker.

[0090] Description of Reference Numerals

[0091] 1: Microphone unit

[0092] 2: Different devices

[0093] 2A: Second microphone

[0094] 10: First Microphone

[0095] 11: Sound data acquisition unit

[0096] 12: Voice data registration department

[0097] 13: Evaluation audio data acquisition unit

[0098] 14: Control section

[0099] 15: Comparison result output section

[0100] 20: Evaluation Department

[0101] 21: Correction Department

[0102] 30: Test sound data acquisition unit

[0103] 31: Computing Department

[0104] 40: Parameter change section.

Claims

1. A microphone unit capable of determining whether a sound input to a first microphone is a sound of an expected speaker, the microphone unit comprising: a sound data acquisition unit that acquires sound as sound data; a voice data registering unit that registers voice data for comparison in which feature points are extracted from the voice data; an evaluation sound data acquisition unit that acquires the sound input to the first microphone as evaluation sound data; a comparison unit that compares whether a speaker of the voice based on the evaluation voice data is the same as a speaker of the voice based on the comparison voice data based on the comparison voice data, based on the comparison voice data and feature points extracted from the evaluation voice data; as well as a comparison result output unit, outputting the comparison result of the comparison unit, The audio data for comparison is produced by a device different from the device equipped with the first microphone, and the device equipped with the first microphone and the different device exchange the audio data for comparison by wireless communication.

2. The microphone unit according to claim 1, wherein When the comparison unit is in the sleep state, the sleep state is ended using the acquisition of the evaluation audio data by the evaluation audio data acquisition unit as a trigger.

3. The microphone unit according to claim 1 or 2, wherein: The sound data acquired by the sound data acquisition unit is sound input to a second microphone provided in a device different from the device equipped with the first microphone. Also available: an evaluation unit that evaluates a frequency characteristic of the first microphone and a frequency characteristic of the second microphone before inputting sound to both the first microphone and the second microphone; and The correction unit performs correction so that a frequency characteristic of one of the first microphone and the second microphone matches a frequency characteristic of the other microphone.

4. The microphone unit according to claim 1 or 2, wherein: Also available: a test voice data acquisition unit that acquires, after registering the control voice data and before acquiring the evaluation voice data, a voice of a speaker who emits the voice of the control voice data through the first microphone as the test voice data; as well as a calculation unit that calculates a comparison rate of the speaker based on the test voice data while changing a comparison parameter used in the comparison based on the test voice data and the comparison voice data, The comparison unit performs the comparison based on the comparison parameter at the highest comparison rate among the comparison rates calculated by the calculation unit.

5. The microphone unit according to claim 4, wherein The comparison parameter is an amplification factor for amplifying at least one of the test audio data and the comparison audio data.

6. The microphone unit according to claim 4, wherein: The method further includes a parameter changing unit that automatically changes a parameter of the first microphone based on the comparison parameter when a voice based on the evaluation voice data is input to the first microphone.

7. The microphone unit according to claim 1 or 2, wherein: Based on the comparison result of the comparison unit, a speaker of the sound input to the first microphone is identified.

8. The microphone unit according to claim 1 or 2, wherein: The utterance content of the sound input to the first microphone is estimated, and the operation of the device equipped with the first microphone is controlled based on the estimated content.