CN110807182A

CN110807182A - Speech decoding device and method

Info

Publication number: CN110807182A
Application number: CN201910947181.1A
Authority: CN
Inventors: 邱文达; 简硕辰
Original assignee: Individual
Current assignee: Individual
Priority date: 2019-09-30
Filing date: 2019-09-30
Publication date: 2020-02-18
Anticipated expiration: 2039-09-30
Also published as: CN110807182B

Abstract

The invention relates to a voice decoding device and a method thereof, wherein the method comprises the following steps: receiving and judging whether an identity identification data is correct or not; if the identification data is incorrect, displaying a decoding array, wherein the decoding array comprises a plurality of characters, and the positions of the characters are distributed randomly; receiving a digital voice command, wherein the digital voice command comprises a plurality of decoding characters which are arranged in sequence; judging whether the digital voice command conforms to a preset decoding track; the plurality of decoding characters are arranged in a row track in contrast to the decoding array; when the arranging track is the same as the decoding track, unlocking is judged; the invention enables a user to speak the sequence which accords with the preset decoding track by the decoding array which is displayed randomly, so that outsiders are difficult to crack by means of eavesdropping on the password, and a better anti-theft effect is achieved.

Description

Speech decoding device and method

Technical Field

The present invention relates to a decoding device and method, and more particularly, to a voice decoding device and method for preventing a password from being seen and stolen by inputting a number sequence corresponding to a predetermined track through a voice.

Background

The existing anti-theft locks are of various types, such as key locks in the aspect of simulation, safe box type rotary disc locks, rotary padlocks, safety locks in the aspects of fingerprint identification, voiceprint identification, face identification, iris identification and the like in the aspect of biological identification, or coded locks with digital display, and all have anti-theft functions to a certain extent. In general home door locks, a key lock is often used as an anti-theft lock, and a resident can unlock the door to open the door. In addition, the home door lock also commonly uses a biological identification type lock, such as a face identification electronic lock, and a resident can establish a face identification reference file for people who have authority to enter and exit in advance, so that only the people who have authority can enter and exit the home door through face identification; the digital coded lock is also a lock commonly used for a door lock of a home, and a resident can get in and out of the home only by inputting a correct code, so that the digital coded lock is simple and convenient to use.

However, any lock, such as a key lock, a face recognition electronic lock, or a digital combination lock, may be cracked. As for the face recognition electronic lock, the face recognition can be broken through a 3D printing mode at present; the digital combination lock may be recorded when the resident inputs the password or easily cracked by the surrounding outsiders seeing the password, which causes a security problem.

Disclosure of Invention

In order to increase the anti-theft effect and avoid the password from being easily stolen by outsiders, the invention provides a voice decoding device and a voice decoding method, which enable a user to speak a sequence conforming to a preset decoding track through a randomly displayed decoding array so as to achieve a better anti-theft means.

To achieve the above object, the speech decoding method of the present invention comprises the following steps:

receiving and judging whether an identity identification data is correct or not;

if the identification data is incorrect, displaying a decoding array, wherein the decoding array comprises a plurality of characters, and the positions of the characters are distributed randomly;

receiving a digital voice command, wherein the digital voice command comprises a plurality of decoding characters which are arranged in sequence;

judging whether the digital voice command conforms to a preset decoding track, wherein the plurality of decoding characters are arranged on the decoding array in a contrasting manner to form an arrangement track;

when the arrangement track is the same as the decoding track, the digital voice command is judged to be consistent with the decoding track.

Wherein, the identification data is voiceprint data of a user.

The identification data is face data of a user.

Wherein, the decoding array is a nine-square grid.

The arrangement track is formed by connecting a plurality of sequentially arranged decoding characters in series in contrast to the decoding array.

The present invention further provides a speech decoding apparatus capable of executing the speech decoding method, the speech decoding apparatus comprising:

a CPU for randomly generating a decoding array and receiving an ID data and a digital voice command, wherein the digital voice command comprises a plurality of sequentially arranged decoding characters, the decoding array comprises a plurality of characters, and the positions of the characters are randomly distributed;

a voice receiving unit electrically connected to the CPU for receiving the digital voice command;

a display unit electrically connected to the CPU for displaying the decoding array;

a storage unit electrically connected to the CPU for storing a decoding track and a standard identity data;

wherein, the plurality of decoding characters are arranged in contrast with the decoding array to form an arrangement track; when the arrangement track is the same as the decoding track, the digital voice command is judged to be consistent with the decoding track.

The voice receiving unit is used for receiving the identity identification data and transmitting the identity identification data to the central processing unit.

The system further comprises an identification unit which is electrically connected with the central processing unit and is used for identifying the identity of a user and generating the identity identification data.

Wherein, the identification data is voiceprint data of a user.

The identification data is face data of a user.

The invention presets the decoding track, and the user can recite the sequence of the randomly displayed decoding array according with the shape of the decoding track in sequence to finish decoding. Because the numbers of the decoding array are randomly arranged, the number series of voice commands spoken by a user every time are different, and a person with worry cannot input the number series of voice commands according to a test record or an overheard number series of voice commands, so that the decoding fails, and the decoding method is safer. Furthermore, the digital voice command of the invention can be used as a second defense line for identity recognition. When the sound emitted by the user is slightly different from the sound of the normal vocal cords due to cold or vocal cords injury, the central processing unit can not judge the identity of the user completely correctly, the display array is further displayed, so that the user can further confirm the identity, and the dilemma that the user can not unlock due to identity identification failure is avoided.

The invention is described in detail below with reference to the drawings and specific examples, but the invention is not limited thereto.

Drawings

FIG. 1: the invention discloses a circuit block diagram of a voice decoding device.

FIG. 2: the speech decoding device of the invention displays a first schematic diagram of a digital Sudoku.

FIG. 3: the steps of the present invention are flow diagrams.

FIG. 4: the invention discloses a face recognition action schematic diagram.

FIG. 5: the decoding trace input action schematic diagram of the invention.

FIG. 6: the speech decoding device of the invention displays a second schematic diagram of a digital Sudoku.

FIG. 7: the speech decoding device of the invention displays the third schematic diagram of the digital Sudoku.

Detailed Description

The invention relates to a voice decoding method, which enables a user to speak a sequence which accords with a specific arrangement mode to a voice decoding device to finish decoding. The voice decoding device can be arranged outside a home door and is responsible for managing entrance guard, and when decoding is successful, the home door is opened to allow a user to come in and go out, so that outsiders are prevented from randomly entering the room.

Referring to fig. 1 and 2, in a first preferred embodiment of the speech decoding apparatus 10, the speech decoding apparatus 10 includes a central processing unit 11, a speech receiving unit 13, a display unit 15 and a storage unit 17. The cpu 11 is configured to randomly generate a decoding array 151 and receive an id data, wherein the decoding array 151 includes a plurality of words, and the positions of the plurality of words are randomly distributed. The decoding array 151 may be a square grid such as a squared grid, a sixteen grid, a twenty-five grid, etc., a square grid such as a twelve grid, a twenty grid, etc., or a honeycomb grid, and the following description takes the squared grid as an example, but the various embodiments of the present invention are not limited to the squared grid.

The voice receiving unit 13 is electrically connected to the cpu 11 for receiving the identification data and transmitting the identification data to the cpu 11, and receiving a digital voice command, wherein the identification data may be user voiceprint data, the digital voice command includes a plurality of sequentially arranged decoded characters, the plurality of decoded characters may be sequentially arranged on the decoded array 151 to form an arrangement track, the arrangement track is a track formed by sequentially connecting the plurality of decoded characters in series with respect to the decoded array 151, for example, if the plurality of decoded digits are sequentially 1, 4, 7, 8, 9, 6, 3, the decoded characters are connected in series on the decoded array (digital grid) 151 to obtain a track of "ㄩ" shape.

The display unit 15 is electrically connected to the cpu 11 for displaying the decoding array 151. The display unit 15 can also provide a touch function for a user to operate the speech decoding device 10 by touching the display unit 15.

The storage unit 17 is electrically connected to the cpu 11 and configured to store a predetermined decoding track and a standard identification data, wherein the decoding track and the standard identification data can be set by a user; the decoding track corresponds to the digital voice command. For example, the decoding trajectory may be a zigzag symbol, ㄩ -shaped symbol, or more complex symbol; the standard identification data may be a reference voiceprint of the user. When the voice receiving unit 13 receives the digital voice command, wherein the voice decoding device 10 unlocks the door when the arrangement track formed by the plurality of digits on the decoding array 151 conforms to the decoding track. In addition, a decoding track corresponds to a standard identification data, that is, each authorized user has a set of decoding tracks separately.

In the second preferred embodiment of the voice decoding apparatus, the apparatus further comprises an identification unit 19, wherein the identification unit 19 is electrically connected to the central processing unit 11 for identifying the identity of the user and generating the identity identification data. In a preferred embodiment, the recognition unit 19 can be a face recognizer, which scans the face of the user to obtain a face data. In a preferred embodiment of the identification unit 19, the standard identification data of the storage unit 17 can be the face of the user.

With the above elements in mind, the process flow of the present invention is described below. The steps of the present invention can be divided into two stages, the first stage is to identify the identity of the user, and the second stage is to determine whether the arrangement track formed by the characters inputted by the user through the voice receiving unit 13 is correct. The first stage is explained first.

Please refer to fig. 3, S11: receiving and judging whether the identification data is correct or not. Receiving the voiceprint data of the user as the identification data through the voice receiving unit 13, or receiving the face data of the user through the identification unit 19; taking fig. 4 as an example, the face data of the user is received. When the identification data is not consistent with the standard identification data, the next step is carried out.

And a second stage:

please refer to fig. 5, S12: if the identification data is incorrect, displaying the decoding array 151 for the user to view the number of the decoding array 151; the decoding array comprises the plurality of words, wherein the positions of the plurality of words are randomly distributed.

For example, referring to fig. 2, when the predetermined decoding trace is "Z", the user observes which combinations of digits in the decoding array 151 correspond to the decoding trace being "Z", and for example, as shown in fig. 2, the words corresponding to the decoding trace being "Z" are "1, 2, 3, 5, 7, 8, 9", and the user sequentially speaks the digital voice command "1, 2, 3, 5, 7, 8, 9" for the voice receiving unit 13 to receive.

S13: the digital voice instruction is received. The voice receiving unit 13 receives the digital voice command spoken by the user according to the number of the decoding array 151.

S14: and judging whether the digital voice command conforms to the decoding track. If the two are matched, the voice decoding device 10 unlocks the door; if they do not match, the speech decoding device 10 does not unlock the door. For example, in fig. 2, when the decoding trace is "Z", it can be determined that the trace matches the sequence of "1, 2, 3, 5, 7, 8, 9" as long as the user says in sequence that the trace matches the decoding trace "Z"; if someone else speaks a sequence other than the above, it is determined that the sequence does not match.

Furthermore, the CPU 11 randomly generates the decoding array 151 such that the number arrangement of the decoding array 151 is different each time the lock is unlocked. For example, the decoding array 151 is shown in FIG. 2 for one unlocking; at the next unlocking time, the decoding array 151 is randomly displayed as shown in fig. 6, and it can be seen that the decoding array 151 of fig. 2 is different from the decoding array 151 of fig. 6 in the number arrangement. However, no matter what kind of the decoding array 151 is encountered by the user, it is determined that the sequence corresponding to the decoding track "Z" is the digital voice command. As shown in the decoding array 151 of fig. 6, when the user speaks the digital voice command "7, 9, 6, 8, 3, 2, 1" in sequence, the shape of the trajectory formed by the sequence of "7, 9, 6, 8, 3, 2, 1" matches the decoding trajectory "Z", i.e. the unlocking is determined to be successful. Referring to fig. 7, if a third person writes a number sequence "1, 2, 3, 5, 7, 8, 9" and says "1, 2, 3, 5, 7, 8, 9" in sequence during a certain unlocking process, and the number arrangement of the decoding array 151 changes, taking fig. 7 as an illustrative example, a track "ㄩ" is arranged according to the number sequence "1, 2, 3, 5, 7, 8, 9" of the decoding array 151, and the track "ㄩ" does not coincide with the decoding track "Z", so that the door lock cannot be successfully unlocked, and the anti-theft effect is achieved.

Therefore, the invention can be applied to further start the function of digital voice command identification when the identification data identification has doubt. When the user is damaged in the vocal cords or has an injured face, the cpu 11 cannot recognize the identification data by a percentage, and the cpu 11 starts to display the decoding array 151, so that the user can obtain a chance of secondary recognition by the digital voice command recognition method and still enter the home by the digital voice command recognition method. Taking voice recognition as an illustrative example, 100 voice recognition data are generated when a user speaks, if the vocal cords of the user are normal and only 90 voice recognition data are correctly recognized, the user can unlock the phone correctly (the recognition rate is 90%), and the central processing unit 11 can unlock the phone after successfully recognizing 90 voice recognition data; if the user's vocal cords are damaged, so that the CPU 11 can only successfully recognize 60 voice recognition data, the CPU 11 will activate the display of the decoding array 151 to further confirm whether the user's identity is correct.

The present invention is capable of other embodiments, and various changes and modifications may be made by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A speech decoding method, comprising the steps of:

judging whether the digital voice command conforms to a preset decoding track, wherein the plurality of sequentially arranged decoding characters are arranged to form an arrangement track by contrasting with the decoding array;

when the arrangement track is the same as the preset decoding track, the digital voice command is judged to be consistent with the preset decoding track.

2. The speech decoding method of claim 1 wherein the identification data is a user's voiceprint data.

3. The speech decoding method of claim 1 wherein the identification data is facial data of a user.

4. The speech decoding method according to any one of claims 1 to 3, wherein the decoding array is a Sudoku.

5. The speech decoding method of claim 4 wherein the arrangement track is a track formed by sequentially connecting the plurality of sequentially arranged decoded characters in series against the decoding array.

6. An apparatus for speech decoding, comprising:

wherein, the plurality of sequentially arranged decoding characters are arranged to form an arrangement track in contrast to the decoding array; when the arrangement track is the same as the decoding track, the digital voice command is judged to be consistent with the decoding track.

7. The apparatus of claim 6 wherein the voice receiving unit is configured to receive the identification data and transmit the identification data to the CPU.

8. The apparatus of claim 6 further comprising an identification unit electrically connected to the CPU for identifying a user's identity and generating the identification data.

9. The speech decoding method according to claim 7 or 8, wherein the identification data is a user's voiceprint data.

10. The speech decoding method of claim 8 wherein the identification data is facial data of a user.