CN104361886B - Speech ciphering method and device and voice decryption method and device - Google Patents

Abstract

The embodiment of the present invention provides a kind of speech ciphering method and device and voice decryption method and device.The speech ciphering method includes：The character in the voice for encryption is recognized, encrypted characters sequence is obtained；Extract the characteristic parameter of the voice for being used to encrypt；It is encrypted with the obtained encrypted characters sequence pair of identification information to be encrypted, and preserve the encrypted characters sequence and the characteristic parameter, for when decryption obtains described information, the characteristic parameter with decryption character string and for the voice of decryption is matched.Therefore decoding difficulty is added, improves the security of encryption information.

Description

Voice encryption method and device and voice decryption method and device

technical field

The embodiment of the present invention relates to the field of information security technologies, and in particular to a voice encryption method and device, and a voice decryption method and device.

Background technique

Speech encryption technology is a commonly used information encryption technology. It mainly recognizes characters from the input voice through speech recognition to obtain an encrypted character sequence, and then encrypts the encrypted information with the encrypted character sequence through an encryption algorithm to ensure the above-mentioned Information Security. Correspondingly, when decrypting encrypted information, voice recognition is performed on the voice used for decryption through voice recognition, characters are recognized from the voice used for decryption, and a sequence of decrypted characters is obtained. Afterwards, match the decrypted character sequence with the encrypted character sequence, and if the match is successful, use the decrypted character sequence or the encrypted character sequence to decrypt the encrypted information through the decryption algorithm corresponding to the above encryption algorithm to obtain the protected information. It can be seen that the existing voice encryption and decryption technology makes the encrypted information easier to decipher, resulting in reduced information security.

Contents of the invention

Embodiments of the present invention provide a voice encryption method and device, and a voice decryption method and device, so as to increase the difficulty of deciphering and improve the security of information.

In the first aspect, the embodiment of the present invention provides a voice encryption method, including:

Recognize the characters in the speech used for encryption to obtain the encrypted character sequence;

Extracting the characteristic parameters of the speech used for encryption;

Using the identified encrypted character sequence to encrypt the information to be encrypted, and saving the encrypted character sequence and the characteristic parameters, for use in decrypting the decrypted character sequence and the voice used for decryption when the information is obtained by decrypting The characteristic parameters are matched.

In a second aspect, the embodiment of the present invention also provides a voice encryption device, including:

The speech recognition module is used for recognizing the character in the speech that is used for encrypting, obtains the encrypted character sequence;

A feature extraction module, used to extract the feature parameters of the speech used for encryption;

The encryption module is used to encrypt the information to be encrypted by using the encrypted character sequence recognized by the speech recognition module, and save the encrypted character sequence and the characteristic parameters, so that when the information is obtained by decryption, It is matched with the decrypted character sequence and the characteristic parameters of the speech used for decryption.

The voice encryption method and device provided by the embodiments of the present invention extract and store the characteristic parameters of the voice used to encrypt information during the voice encryption process, so that when the information is obtained by decryption, it can be used for decrypting the character sequence and used for decryption. The feature parameters of the speech are matched, so that when a decryptor with bad intentions tries to decipher the encrypted information, not only the characters carried by the speech must be consistent, but also the speech features must be matched successfully to achieve the purpose of deciphering, thereby increasing the The difficulty of deciphering improves the security of encrypted information.

In a third aspect, the embodiment of the present invention also provides a voice decryption method, including:

Recognize the voice used to decrypt the encrypted information to obtain the decrypted character sequence;

Extracting the characteristic parameters of the speech used for decryption;

Matching the decrypted character sequence and the characteristic parameters of the speech used for decryption with the saved encrypted character sequence and characteristic parameters respectively, the encrypted character sequence and characteristic parameters are the encryption of the speech encrypted by the encrypted information Character sequences and feature parameters;

After the matching is successful, the encrypted information is decrypted by using the decrypted character sequence or the encrypted character sequence.

In a fourth aspect, the embodiment of the present invention also provides a voice decryption device, including:

The voice recognition module is used to identify the voice used to decrypt the encrypted information to obtain the decrypted character sequence;

Feature extraction module, used for extracting the feature parameter of the speech that is used for decrypting;

A matching module, configured to match the decrypted character sequence and the characteristic parameters of the speech used for decryption with the stored encrypted character sequence and characteristic parameters, the encrypted character sequence and characteristic parameters are the encrypted information The encrypted character sequence and characteristic parameters of the encrypted voice;

The decryption module is configured to use the decrypted character sequence or the encrypted character sequence to decrypt the encrypted information after the matching is successful.

The speech decryption method and device provided by the embodiments of the present invention extract the characteristic parameters of the speech during the decryption process, and when the decrypted character sequence and the extracted characteristic parameters are successfully matched with the stored encrypted character sequence and characteristic parameters respectively, The encrypted information is decrypted by using the decrypted character sequence or the encrypted character sequence. Compared with the prior art, only matching the decrypted character sequence and the encrypted character sequence increases the difficulty of deciphering and improves the security of the information.

Description of drawings

In order to illustrate the present invention more clearly, the accompanying drawings that need to be used in the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art , on the premise of not paying creative labor, other drawings can also be obtained based on these drawings.

Fig. 1a is a flowchart of a voice encryption method provided by Embodiment 1 of the present invention;

Fig. 1 b is a schematic diagram of the waveform of the voice decibel value in the voice encryption method provided by the embodiment of the present invention;

Fig. 2 is the flowchart of a kind of speech encryption method that the embodiment 2 of the present invention provides;

FIG. 3 is a schematic structural diagram of a voice encryption device provided by Embodiment 3 of the present invention;

FIG. 4 is a flow chart of a speech decryption method provided in Embodiment 4 of the present invention;

FIG. 5 is a schematic structural diagram of a speech decryption device provided by Embodiment 5 of the present invention.

detailed description

In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Example. It can be understood that the specific embodiments described here are only used to explain the present invention, rather than limit the present invention. Based on the embodiments of the present invention, all those skilled in the art can obtain without creative work. Other embodiments all belong to the protection scope of the present invention. In addition, it should be noted that, for the convenience of description, only parts related to the present invention are shown in the drawings but not all content.

Embodiment one

Please refer to FIG. 1a, which is a flowchart of a voice encryption method provided by Embodiment 1 of the present invention. The method of the embodiment of the present invention can be executed by a voice encryption device implemented in hardware and/or software. The implementation device is typically configured in a terminal capable of providing voice encryption services, such as a smart phone, a tablet computer, etc. various mobile terminals.

The method includes: operation 11-operation 13.

In operation 11, characters in the speech for encryption are identified to obtain an encrypted character sequence.

This operation specifically uses speech recognition technology to recognize and obtain an encrypted character sequence from the speech used for encryption.

Wherein, the voice used for encryption refers to the voice for encrypting the information to be encrypted, which can be the voice input by the user through the microphone in the terminal capable of providing voice encryption service, or the voice stored locally in the terminal, or It is a voice clip downloaded in real time. Preferably, the voice input by the user through the microphone in the terminal capable of providing voice encryption services is adopted. For the information to be encrypted, the user can input an appropriate voice according to his own needs, which is convenient for the user to remember, and the content of the input voice can be controlled by the user independently. Increased flexibility.

The voice used for encryption may be in Chinese or in other forms such as English, which is not limited in this embodiment, as long as it can be correctly recognized by the terminal providing the voice encryption service.

The encrypted character sequence refers to a character sequence formed from characters recognized from the encrypted speech according to the sequence of recognizable phonemes in the encrypted speech. For example, the voice used for encryption is "open the door", and the characters "kai" and "men" are recognized, and the characters "kai" and "men" are used in the order of recognizable phonemes in the voice used for encryption to form an encrypted The character sequence "kaimen".

In operation 12, the characteristic parameters of the speech used for encryption are extracted.

Voice features include priorities, etc. For example, A usually speaks in a deep and slow voice, and B usually speaks quickly and with a high voice. That is to say, different people have different voice features. If these features are extracted and saved, and used for decryption Matching with the characteristics of the voice that is trying to be decrypted can ensure that only the person who encrypts can decrypt the encrypted information, improving the security of the information.

In operation 13, encrypt the information to be encrypted by using the identified encrypted character sequence, and save the encrypted character sequence and the characteristic parameters, so as to be used for combining with the decrypted character sequence and the used Match the feature parameters of the decrypted speech.

This operation not only includes the operation of encrypting the information to be encrypted with the identified encrypted character sequence, but also includes the operation of saving the encrypted character sequence and the characteristic parameters, both of which are not limited in order, and can also be executed in parallel. When using the identified encrypted character sequence to encrypt the information to be encrypted, the encryption algorithm used may be AES256, etc., which is not limited here.

Wherein, the information to be encrypted may include various types such as documents, video files, audio files, or pictures, or may be a combination of the aforementioned types.

Wherein, storing the encrypted character sequence and the characteristic parameters is specifically used to match the decrypted character sequence and the characteristic parameters of the speech used for decryption when decrypting the information, in other words, when the encrypted information is When decrypting, it is not only necessary to match the decrypted character sequence with the saved encrypted character sequence, but also to match the characteristic parameters of the speech used for decryption with the saved characteristic parameters of the speech used for encryption.

In the technical solution of this embodiment, by extracting and storing the characteristic parameters of the speech used for encrypting information during the speech encryption process, when the information is obtained by decryption, it can be used for decrypting the character sequence and the characteristic parameters of the speech used for decryption. Matching is carried out, so that when a decryptor with bad intentions tries to decipher the encrypted information, not only the characters carried by the voice must be consistent, but also the characteristics of the voice must be matched successfully to achieve the purpose of deciphering, thereby increasing the difficulty of deciphering and improving the Security of Encrypted Information.

It should be noted that, in this embodiment, operation 11 and operation 12 are independent of each other, and there is no necessary order restriction, and may be performed in a reverse order to that of this embodiment, or may be performed in parallel.

It should also be noted that the extraction of the characteristic parameters of the speech used for encryption may include: extracting at least one of the duration, speech rate and decibel value of the speech used for encryption.

Extracting the decibel value of the speech used for encryption may be to obtain the decibel value at each time point after the speech used for encryption is recognized; or it may also be to obtain the decibel value of each character in the encrypted character sequence decibel value. As shown in Figure 1b, the decibel value is a curve on the time axis, and the decibel value is different at different time points. Convert the waveform diagram in Figure 1b into data, which can be saved as an array, that is, the decibel value at each time point. If the time for saying "z_h_i__m_a_______k_a_i___m_e_n" (see the description in the following embodiment two) is 3 seconds, that is to say, there are 30 0.1 seconds, then the decibel value can be expressed as an array: [30,20,40,30, 35,30,20,10,50,50,50,40,25,15,5,0,0,0,0,30,35,35,35,25,15,35,35,35,35, 35]. Encrypted character sequences, speech rate values and decibel values can be saved separately to a file. Even for the same encrypted character sequence, due to the difference in the way the encryptor speaks, the decibel values of the speech used for encryption input by different encryptors are usually inconsistent, so the encrypted character sequence and the decibel of the speech used for encryption are saved when encrypting When decrypting, even if the decrypted character sequence matches the encrypted character sequence successfully, the decibel value of the encrypted voice can still increase the difficulty of deciphering, thereby improving the security of encrypted information.

The more feature parameters extracted, the longer the time-consuming, but the more difficult it is to decipher the encrypted information, the higher the security of the encrypted information; the fewer the feature parameters extracted, the shorter the time-consuming, but the encrypted information The lower the difficulty of deciphering the information, the lower the security of the encrypted information. Therefore, choosing to extract appropriate feature parameters can take into account the encryption efficiency and the security of the encrypted information.

The implementation manner of extracting the speech rate value of the speech used for encryption will be described below through specific embodiments.

Embodiment two

Please refer to FIG. 2 , which is a flowchart of a voice encryption method provided by Embodiment 2 of the present invention. On the basis of the foregoing embodiments, this embodiment provides a preferred solution for the operation of extracting the characteristic parameters of the speech used for encryption, and specifically provides a preferred solution for extracting the speech rate value of the speech used for encryption.

The preferred method includes: operation 21-operation 22.

In operation 21, the time interval between two adjacent characters in the encrypted character sequence is obtained.

In this embodiment, by acquiring the time interval between two adjacent characters in the encrypted character sequence, the obtained time interval is used as the speech rate value of the speech used for encryption. In other words, based on the pause between two adjacent characters in the encrypted character sequence, characterize the speech rate value of the speech used for encryption. For example, when a person wants to encrypt information, he first presses the virtual microphone button on the screen, speaks to the person, such as saying "Open Sesame", raises his hand after speaking, and ends speaking. There are various ways of starting and ending. So, there is no limit here.

The encryption device will perform speech recognition on "Open Sesame", and analyze the distribution of the following characters on the time axis:

#______z_h_i__m_a_______k_a_i___m_e_n____#

Among them, the first "#" indicates the start, and the second "#" indicates the end; the time taken by each "_" or letter is 0.1 second. The waiting time before z and after n is ignored, and only the part between z and n is analyzed to obtain the time interval between two adjacent letters, that is, the pause time difference between every two phonemes.

Due to the difference in the way the encryptor speaks, the pause between two adjacent characters in the speech input by different encryptors for encryption is inconsistent. Therefore, when decrypting, even if the decrypted character sequence matches the encrypted character sequence successfully, it is still possible. According to the matching result of the time interval between two adjacent letters between the encrypted and decrypted voices, the difficulty of deciphering is increased, thereby improving the security of the encrypted information.

In operation 22, the duration of the speech used for encryption is obtained, and the speech rate value of the speech used for encryption is obtained according to the duration and the number of characters in the encrypted character sequence.

For example, "#______z_h_i__m_a_______k_a_i___m_e_n____#" through speech recognition can get the speech duration, 40×0.1 second=4 seconds, and there are 11 characters in total, so the speech rate value can be obtained as 4÷11 seconds.

Due to the difference in speech speed of the encryptors, the duration of the speech used for encryption input by different encryptors is usually inconsistent. Therefore, when decrypting, even if the decrypted character sequence matches the encrypted character sequence successfully, it can still be based on the speech used for encryption and decryption. The matching results of speech rate values between speech rates increase the difficulty of deciphering, thereby further improving the security of encrypted information.

The difference with operation 21 is that operation 21 characterizes the speech rate value of the speech used for encryption based on the pause between two adjacent characters in the encrypted character sequence; and operation 22 is based on the speech rate value of the speech used for encryption and the number of characters in the encrypted character sequence, the speech rate value of the speech used for encryption is characterized by the time consumption of each character.

It should be noted that in this embodiment, operation 21 and operation 22 are independent of each other, and both are preferred implementations of extracting the speech rate value of the speech used for encryption, in other words, extracting the speech rate value of the speech used for encryption The speech rate value of may include at least one of operation 21 and operation 22, and when multiple operations are performed, the order in which they are performed is not limited.

When executed in combination, the more feature parameters extracted, the longer the time-consuming, but the more difficult it is to decipher the encrypted information, the higher the security of the encrypted information; when one of them is executed alone, the extracted feature parameters The less it takes, the shorter the time-consuming, but the lower the difficulty of deciphering the encrypted information, the lower the security of the encrypted information. Therefore, choosing to extract appropriate feature parameters can take into account the encryption efficiency and the security of the encrypted information.

Embodiment three

Please refer to FIG. 3 , which is a schematic structural diagram of a voice encryption device provided by Embodiment 3 of the present invention. The device includes: a voice recognition module 31 , a feature extraction module 32 and an encryption module 33 .

Wherein, recognition module 31 is used for identifying the character in the speech that is used for encrypting, obtains encrypted character sequence; Feature extraction module 32 is used for extracting the characteristic parameter of described speech that is used for encrypting; Encryption module 33 is used for using described speech recognition The encrypted character sequence obtained by module identification is used to encrypt the information to be encrypted, and save the encrypted character sequence and the characteristic parameters, so that when the information is obtained by decryption, the decrypted character sequence and the voice used for decryption The characteristic parameters are matched.

In the technical solution of this embodiment, by extracting and storing the characteristic parameters of the speech used for encrypting information during the speech encryption process, when the information is obtained by decryption, it can be used for decrypting the character sequence and the characteristic parameters of the speech used for decryption. Matching is carried out, so that when a decryptor with bad intentions tries to decipher the encrypted information, not only the characters carried by the voice must be consistent, but also the characteristics of the voice must be matched successfully to achieve the purpose of deciphering, thereby increasing the difficulty of deciphering and improving the Security of Encrypted Information.

In the above solution, the feature extraction module 32 preferably includes at least one of a time extraction submodule, a speech rate extraction submodule and a decibel extraction submodule.

Wherein, the time extraction submodule is used to extract the duration of the speech used for encryption; the speech rate extraction submodule is used to extract the speech rate value of the speech used for encryption; the decibel extraction submodule is used to extract the speech rate used for The decibel level of encrypted speech.

The speech rate extraction submodule is used to perform at least one of the following:

Obtain the time interval between two adjacent characters in the encrypted character sequence;

The duration of the speech used for encryption is obtained, and the speech rate value of the speech used for encryption is obtained according to the duration and the number of characters in the encrypted character sequence.

The decibel extraction submodule is used to:

Obtain the decibel value at each time point after the voice used for encryption is recognized; or

Used to obtain the decibel value of each character in the encrypted character sequence.

The speech encryption device provided in the embodiment of the present invention can execute the speech encryption method provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

Embodiment Four

Please refer to FIG. 4 , which is a flowchart of a speech decryption method provided by Embodiment 4 of the present invention. The method of the embodiment of the present invention can be executed by a voice decryption device implemented in hardware and/or software, and the implementation device is typically configured in a terminal capable of providing voice decryption services, such as a smart phone, a tablet computer, etc. various mobile terminals.

The method includes: operation 41 to operation 44 .

In operation 41, the speech used for decrypting the encrypted information is recognized to obtain a sequence of decrypted characters.

This operation specifically uses speech recognition technology to recognize and obtain a decrypted character sequence from the speech used to decrypt the encrypted information.

The decrypted character sequence refers to the character sequence formed from the characters recognized from the speech that decrypts the encrypted information, according to the sequence of identifiable phonemes in the speech used to decrypt the encrypted information.

In operation 42, feature parameters of the speech for decryption are extracted.

This operation is the same as the method of extracting the characteristic parameters of the encrypted speech in the above encryption embodiment, and will not be repeated here.

In operation 43, the decrypted character sequence and the characteristic parameters of the speech used for decryption are matched with the stored encrypted character sequence and characteristic parameters, and the encrypted character sequence and characteristic parameters are used to encrypt the encrypted information. The encrypted character sequence and feature parameters of the speech.

This operation not only includes matching the decrypted character sequence with the saved encrypted character sequence of the speech encrypted by the encrypted information, but also includes matching the characteristic parameters of the speech used for decryption with the saved pair of the encrypted speech The encrypted information is used to match the characteristic parameters of the encrypted voice.

Wherein, the encrypted character sequence and characteristic parameters of the voice encrypted by the encrypted information can be stored in the terminal.

Wherein, the decrypted character sequence is matched with the stored encrypted character sequence of the speech encrypted by the encrypted information, preferably the decrypted character sequence is completely matched with the encrypted character sequence to increase the difficulty of deciphering, Improve the security of encrypted information.

Wherein, matching the characteristic parameter of the speech used for decryption with the saved characteristic parameter preferably includes: judging whether the value of the characteristic parameter of the speech used for decryption is within the preset value of the saved characteristic parameter Within the deviation range, if the value of the characteristic parameter of the speech used for decryption is within the preset deviation range, the matching is successful. In other words, as long as the characteristic parameters of the speech used for decryption are fuzzy matched with the saved characteristic parameters, the reason for fuzzy matching is that since the same encryptor speaks the same encrypted speech each time, although the encrypted character sequence is consistent, the Since the way of speaking cannot be exactly the same every time, the characteristic parameters cannot be completely consistent. Therefore, in order to improve the decryption efficiency, as long as the characteristic parameters of the speech used for decryption are roughly the same as the saved characteristic parameters, it is not necessary to be exactly the same. The similarity of the fuzzy matching can be realized by presetting the deviation range of the value of the saved characteristic parameter.

Assume that the matching duration is described by taking the encrypted voice "Open Sesame" as an example. Get "z_h_i__m_a_______k_a_i___m_e_n" (see the description in the above-mentioned embodiment two) and a duration of 3 seconds through speech recognition, thereby obtaining an encrypted character sequence -- "zhimakaimen", save "zhimakaimen" and a duration of 3 seconds. When decrypting, after the decrypted character sequence obtained by speech recognition is "zhimakaimen", it is also necessary to judge whether the duration of the speech used for decryption is within the range of 3 seconds ± 3 × 20%, where ± 3 × 20% is the preset deviation range , depending on actual needs, and there is no limit here. If the duration of the speech used for decryption is within the range of 3 seconds±3×20%, that is, between 2.4 seconds and 3.6 seconds, the duration matching is successful.

Still taking the voice "Open Sesame" used for encryption as an example (see the description in the second embodiment above), the time interval between two adjacent characters in the matching encryption and decryption character sequence will be described. , from the first character "z" to the eleventh character "n", the time intervals between two adjacent characters are 0.1 second, 0.1 second, 0.2 second, 0.1 second, 0.7 second, 0.1 second, 0.1 second, 0.3 seconds, 0.1 seconds and 0.1 seconds. Still assuming that the preset deviation range is ±20% of the time interval between two adjacent characters in the encrypted character sequence, when the decrypted character sequence matches the encrypted character sequence successfully, if every two adjacent characters in the decrypted character sequence The difference between the time interval between and the time interval between corresponding adjacent characters in the encrypted character sequence, within the range of ±20% of the time interval between corresponding adjacent characters in the encrypted character sequence, such as decrypted characters The time interval between every two adjacent characters in the sequence is 0.09 seconds, 0.11 seconds, 0.17 seconds, 0.1 seconds, 0.57 seconds, 0.09 seconds, 0.1 seconds, 0.25 seconds, 0.11 seconds and 0.09 seconds according to the sequence of characters, then The match was successful.

Similarly, for the decibel value at each time point after the speech used for encryption is recognized, a corresponding preset deviation range can also be set, and it is judged that the decibel value at each time point after the speech used for decryption is recognized is consistent with Whether the deviation between the decibel values at the corresponding time points in the voice used for encryption is within the preset deviation range, if so, the matching is successful.

Alternatively, it is also possible to set a corresponding preset deviation range only for the decibel value corresponding to each character in the encrypted character sequence, and determine the difference between the decibel value of each character in the decrypted character sequence and the corresponding character in the encrypted character sequence. Whether the deviation is within the preset deviation range, if so, the matching is successful.

Since the feature parameters of the speech used for decryption need to be matched with the preserved feature parameters for fuzzy matching during decryption, even if the encrypted character sequence is exposed, if the feature parameter matching fails, the protected information cannot be decrypted, so , not only improves the security of the information, but also can openly use the encrypted speech text as a prompt, so as to prevent the user who protects the information through the voice encryption technology from forgetting the password.

In operation 44, after the matching is successful, the encrypted information is decrypted by using the decrypted character sequence or the encrypted character sequence.

In the technical solution of this embodiment, by extracting the characteristic parameters of the speech during the decryption process, and when the decrypted character sequence and the extracted characteristic parameters are successfully matched with the preserved encrypted character sequence and the characteristic parameters respectively, the decrypted character sequence or The encrypted character sequence decrypts the encrypted information. Compared with the prior art, only matching the decrypted character sequence and the encrypted character sequence increases the difficulty of deciphering and improves the security of the information.

It should be noted that in this embodiment, operation 41 and operation 42 are independent of each other, and there is no necessary order restriction, and may be performed in a reverse order to that of this embodiment, or may be performed in parallel.

It can be known from the foregoing first and second embodiments that extracting the characteristic parameters of the speech used for encryption preferably includes: extracting at least one of the duration, speech rate and decibel value of the speech used for encryption. Correspondingly, extracting the characteristic parameters of the speech for decryption preferably includes: extracting at least one of the duration, speech rate and decibel value of the speech for decryption. Which type of characteristic parameter of the speech used for decryption is specifically extracted depends on the type of the saved characteristic parameter of the speech encrypted by the encrypted information.

Wherein, extracting the speech rate value of the speech used for decryption may include obtaining the time interval between two adjacent characters in the decrypted character sequence; it may also include obtaining the duration of the speech used for decryption, according to The speech rate value of the speech used for decryption is obtained by the duration and the number of characters in the decrypted character sequence.

Similar to the foregoing second embodiment, details are not repeated here.

Wherein, extracting the decibel value of the speech used for decryption may be to obtain the decibel value at each time point after the speech used for decryption is recognized; or it may also be to obtain each decibel value in the decrypted character sequence The decibel value of the character.

Embodiment five

Please refer to FIG. 5 , which is a schematic structural diagram of a speech decryption device provided by Embodiment 5 of the present invention. The device includes: a speech recognition module 51 , a feature extraction module 52 , a matching module 53 and a decryption module 54 .

Wherein, the identification module 51 is used to identify the speech used to decrypt the encrypted information to obtain the decrypted character sequence; the characteristic parameter extraction module 52 is used to extract the characteristic parameter of the speech used for decryption; the matching module 53 is used to use the The decrypted character sequence and the characteristic parameter of the speech used for decryption are respectively matched with the stored encrypted character sequence and characteristic parameter, and the encrypted character sequence and characteristic parameter are the encrypted character sequence of the speech encrypted by the encrypted information and characteristic parameters; the decryption module 54 is used to decrypt the encrypted information with the decrypted character sequence or the encrypted character sequence after the matching is successful.

In the technical solution of this embodiment, by extracting the characteristic parameters of the speech during the decryption process, and when the decrypted character sequence and the extracted characteristic parameters are successfully matched with the preserved encrypted character sequence and the characteristic parameters respectively, the decrypted character sequence or The encrypted character sequence decrypts the encrypted information. Compared with the prior art, only matching the decrypted character sequence and the encrypted character sequence increases the difficulty of deciphering and improves the security of the information.

In the above solution, the feature extraction module 52 preferably includes at least one of a time extraction submodule, a speech rate extraction submodule and a decibel extraction submodule.

Wherein, the time extraction submodule is used to extract the duration of the speech used for decryption; the speech rate extraction submodule is used to extract the speech rate value of the speech used for decryption; the decibel extraction submodule is used to extract the speech rate used for decryption The decibel value of the decrypted speech.

The speech rate extraction submodule is preferably used to perform at least one of the following:

Obtain the time interval between two adjacent characters in the decrypted character sequence;

The duration of the speech used for decryption is obtained, and the speech rate value of the speech used for decryption is obtained according to the duration and the number of characters in the decrypted character sequence.

The decibel extraction sub-module is preferably used to: obtain the decibel value at each time point after the speech used for decryption is recognized; or obtain the decibel value of each character in the decrypted character sequence.

In the above solution, the matching module 53 can be specifically used to: determine whether the value of the characteristic parameter of the speech used for decryption is within the preset deviation range of the value of the characteristic parameter saved, if the If the value of the characteristic parameter of the decrypted speech is within the preset deviation range, the matching is successful.

The speech decryption device provided in the embodiment of the present invention can execute the speech decryption method provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; preferred implementations in the examples are not to limit them, and for those skilled in the art, this The invention is capable of various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims (10)

1. a kind of speech ciphering method, it is characterised in that including：

The character in the voice for encryption is recognized, encrypted characters sequence is obtained；

Extract the characteristic parameter of the voice for being used to encrypt；

It is encrypted with the obtained encrypted characters sequence pair of identification information to be encrypted, and preserves the encrypted characters sequence And the characteristic parameter, for when decryption obtains described information, the spy of the voice with decryption character string and for decryption Parameter is levied to be matched；

Wherein, the characteristic parameter of voice encrypted is used for described in the extraction, including：

Extract at least one in duration, pace value and the decibel value of the voice for being used to encrypt；

At least one of wherein, for the pace value for the voice encrypted described in extracting, including following：

Obtain the time interval between two neighboring character in the encrypted characters sequence；

The duration of the voice for being used to encrypt is obtained, according to the quantity of character in the duration and the encrypted characters sequence, Obtain the pace value of the voice for being used to encrypt.

2. according to the method described in claim 1, it is characterised in that extract the decibel value of the voice for being used to encrypt, including：

Obtain the decibel value on each time point after the voice for being used to encrypt is identified；Or

Obtain the decibel value of each character in the encrypted characters sequence.

3. a kind of voice encryption device, it is characterised in that including：

Sound identification module, for recognizing the character in the voice for encryption, obtains encrypted characters sequence；

Characteristic extracting module, the characteristic parameter for extracting the voice for being used to encrypt；

Encrypting module, for recognizing that obtained encrypted characters sequence pair information to be encrypted is entered with the sound identification module Row encryption, and the encrypted characters sequence and the characteristic parameter are preserved, for when decryption obtains described information, with decryption Character string and characteristic parameter for the voice of decryption are matched；

Wherein, the characteristic extracting module includes at least one following submodule：

Time extracting sub-module, the duration for extracting the voice for being used to encrypt；

Word speed extracting sub-module, the pace value for extracting the voice for being used to encrypt；

Decibel extracting sub-module, the decibel value for extracting the voice for being used to encrypt；

Wherein, the word speed extracting sub-module is at least one of following for performing：

Obtain the time interval between two neighboring character in the encrypted characters sequence；

The duration of the voice for being used to encrypt is obtained, according to the quantity of character in the duration and the encrypted characters sequence, Obtain the pace value of the voice for being used to encrypt.

4. device according to claim 3, it is characterised in that the decibel extracting sub-module is used for：

Obtain the decibel value on each time point after the voice for being used to encrypt is identified；Or

Obtain the decibel value of each character in the encrypted characters sequence.

5. a kind of voice decryption method, it is characterised in that including：

Recognize for being encrypted the voice that information is decrypted, obtaining decrypting character string；

Extract the characteristic parameter of the voice for being used to decrypt；

By the characteristic parameter of the decryption character string and the voice for being used to decrypt and the encrypted characters sequence preserved and spy Levy parameter to be matched respectively, the encrypted characters sequence and characteristic parameter are adding to the voice of the encrypted information encryption Close character string and characteristic parameter；

After the match is successful, it is decrypted with information is encrypted described in the decryption character string or the encrypted characters sequence pair；

Wherein, the characteristic parameter of voice decrypted is used for described in the extraction, including：

Extract at least one in duration, pace value and the decibel value of the voice for being used to decrypt.

At least one of wherein, for the pace value for the voice decrypted described in extracting, including following：

Obtain the time interval between two neighboring character in the decryption character string；

The duration of the voice for being used to decrypt is obtained, according to the quantity of character in the duration and the decryption character string, Obtain the pace value of the voice for being used to decrypt.

6. method according to claim 5, it is characterised in that extract the decibel value of the voice for being used to decrypt, including：

Obtain the decibel value on each time point after the voice for being used to decrypt is identified；Or

Obtain the decibel value of each character in the decryption character string.

7. according to any described methods of claim 5-6, it is characterised in that by the characteristic parameter of the voice for being used to decrypt Matched with the characteristic parameter of preservation, including：

Judge the voice for being used to decrypt characteristic parameter value whether the preservation characteristic parameter value it is default partially In poor scope, if the value of the characteristic parameter of the voice for being used to decrypt is in the range of the predetermined deviation, the match is successful.

8. a kind of voice decrypts device, it is characterised in that including：

Sound identification module, is used to, to being encrypted the voice that information is decrypted, obtain decrypting character string for recognizing；

Characteristic extracting module, the characteristic parameter for extracting the voice for being used to decrypt；

Matching module, for by it is described decryption character string and it is described be used for decrypt voice characteristic parameter with preserve encryption Character string and characteristic parameter are matched respectively, and the encrypted characters sequence and characteristic parameter are that the encrypted information is added The encrypted characters sequence and characteristic parameter of close voice；

Deciphering module, after the match is successful, is encrypted with described in the decryption character string or the encrypted characters sequence pair Information is decrypted；

Wherein, the characteristic extracting module includes at least one following submodule：

Time extracting sub-module, the duration for extracting the voice for being used to decrypt；

Word speed extracting sub-module, the pace value for extracting the voice for being used to decrypt；

Decibel extracting sub-module, the decibel value for extracting the voice for being used to decrypt.

Wherein, the word speed extracting sub-module is at least one of following for performing：

Obtain the time interval between two neighboring character in the decryption character string；

The duration of the voice for being used to decrypt is obtained, according to the quantity of character in the duration and the decryption character string, Obtain the pace value of the voice for being used to decrypt.

9. device according to claim 8, it is characterised in that the decibel extracting sub-module is used for：

Obtain the decibel value on each time point after the voice for being used to decrypt is identified；Or

Obtain the decibel value of each character in the decryption character string.

10. according to any described devices of claim 8-9, it is characterised in that the matching module specifically for：

Judge the voice for being used to decrypt characteristic parameter value whether the preservation characteristic parameter value it is default partially In poor scope, if the value of the characteristic parameter of the voice for being used to decrypt is in the range of the predetermined deviation, the match is successful.