KR102140770B1 - Method for unlocking user equipment based on voice, user equipment releasing lock based on voice and computer readable medium having computer program recorded therefor - Google Patents

Abstract

The present invention discloses a user device that performs voice-based unlocking, a method for unlocking a voice-based user device, and a recording medium on which a computer program is recorded. That is, the present invention determines whether the sound information input to the user device is a human voice, and when the sound information is a human voice, generates converted text based on the sound information and compares the converted text and the reference text to the user device You can decide whether to unlock or not. By using this method, it is possible to not only provide a more convenient user-friendly environment such as existing key input, but also increase security.

Description

Method for unlocking user equipment based on voice, user equipment releasing lock based on voice and computer readable medium having a user device performing voice-based unlocking, a method for unlocking a voice-based user device, and a computer program computer program recorded therefor}

The present invention relates to a user device performing unlocking based on voice, a method for unlocking a user device based on voice, and a recording medium on which a computer program is recorded, in particular, unlocking based on the voice of a user input to the user device The present invention relates to a user device, a method for unlocking a user device based on a user voice, and a recording medium on which a computer program is recorded.

2. Description of the Related Art Recently, products such as an automatic interpretation system, a voice recognition telephone, a voice recognition computer, and a voice recognition vehicle have emerged, and the public's interest in voice recognition is increasing. In science fiction films, voice recognition always appears as an essential skill in the future society. Since voice is the most natural and effective means of information exchange for most people, voice recognition technology can be a technology that can provide more convenience in everyday life if activated. The goal of speech recognition technology is to implement a machine capable of receiving speech information and performing appropriate actions and conversations according to the received information. Speech recognition research has been conducted for a long time with the goal of realizing a machine that naturally communicates with humans. Although speech recognition technology that can naturally recognize the voice of an arbitrary speaker under a given environment has not been completed, speech recognition technology is still being applied to various applications.

For example, recent voice-based services have been introduced due to the development of smart phone technology. In order to provide such a voice-based service, an accurate voice recognition technology is required, and a method for adaptively utilizing the voice recognition technology according to the nature and user setting of the service is required.

Korean Patent Publication No. 10-2002-0057262 [Name: Locking device and method of mobile communication terminal using speech recognition]

An object of the present invention is to prevent a third party from unauthorized use of a user device based on a user's voice signal, a user device that unlocks a user device, a method for unlocking the user device, and a recording medium in which a computer program is recorded To provide.

Another object of the present invention is to provide a recording medium in which a user device that utilizes a voice-based interface for unlocking, a user device unlocking method, and a computer program are recorded to improve the convenience of unlocking the user device. have.

Another object of the present invention is to enable the user device and the user device to adaptively set whether to unlock the user device based on the voice signal by setting the parameter value for determining the recognition level by the user. It is to provide a recording medium in which a method of unlocking and a computer program are recorded.

A user device performing unlocking based on voice according to an embodiment of the present invention includes a sound type analyzer configured to determine whether sound information input to a user device is a human voice; If the sound information is a human voice, the match is calculated by generating converted text based on the sound information and comparing the converted text with reference text, which is text information that is a criterion for determining whether to unlock the user device. An individual voice pattern analysis unit implemented to determine whether to unlock the user device based on a result and a preset user setting parameter; And a processor implemented to unlock the user device when the matching result satisfies the threshold of the user setting parameter.

As an example related to the present invention, the matching result includes information on the number of tokens matched between the converted text and the reference text, and information on the matching rate, and the user setting parameter is used to determine whether to unlock the user device. It may be threshold information for a set number of tokens and threshold information for a match rate.

As an example related to the present invention, the number of matching tokens is calculated based on the number of words mapped by performing mapping for each word included in the converted text and the reference text, and the matching rate is based on the number of syllables matching in the mapped word. It can be calculated based on.

As an example related to the present invention, the user device performing unlocking based on voice further includes a user setting unit implemented to set user setting parameters, and the user setting parameters may be set differently according to the degree of security requirements of the user device. have.

As an example related to the present invention, an individual voice pattern analysis unit divides sound information into token units, pre-processes sound information divided into token units based on feature information of the user's voice, and converts the pre-processed sound information into text. It may be implemented to generate converted text, and the feature information of the user voice may be frequency information.

As an example related to the present invention, the converted text may be set and stored as a reference text when a threshold value of a user-set parameter is satisfied.

As an example related to the present invention, the sound type analysis unit may be implemented to determine a human voice when a pattern does not exist in the sound information and is a human voice frequency band.

According to an embodiment of the present invention, a method for unlocking a user voice-based user device includes determining whether sound information input to the user device is human voice; If the sound information is the human voice, generating converted text based on the sound information; Determining whether to unlock the user device based on the matching result calculated by comparing the converted text and the reference text, which is text information that is a judgment criterion for determining whether to unlock the user device, or a user setting parameter. step; And when the matching result satisfies the threshold of the user setting parameter, unlocking the user device.

As an example related to the present invention, the matching result includes information on the number of tokens matched between the converted text and the reference text and information on the matching rate, and the user setting parameter is set to determine whether to unlock the user device. It may be threshold information on the number of tokens and threshold information on the match rate.

As an example related to the present invention, when the sound information is a human voice, generating converted text based on the sound information includes: detecting pitch information from the sound information; Calculating a feature vector from sound information based on the pitch information; Extracting sound information to be converted into converted text among sound information based on the feature vector; And generating extracted sound information as converted text based on a language model modeled based on a linguistic order relationship of words and syllables.

A computer program for performing the method according to the above-described embodiment may be stored in a recording medium on which the computer program according to an embodiment of the present invention is recorded.

The present invention uses a method of recognizing a human voice to unlock a user device, thereby providing a user-friendly environment as compared to a method of unlocking based on an existing key input and enhancing security.

In addition, the present invention not only helps unlocking a user device used by children and persons with disabilities by using a voice-based interface, but also determines whether the user device is unlocked by determining a recognition level according to a setting value of a user setting parameter. There is an effect that can be set adaptively according to the situation.

1 is a conceptual diagram showing a user authentication method based on speech recognition according to an embodiment of the present invention.
2 is a conceptual diagram illustrating a method for unlocking a user device based on speech recognition according to an embodiment of the present invention.
3 is a flowchart illustrating a method of determining a user setting parameter according to an embodiment of the present invention and determining whether to release the locking device.
4 is a conceptual diagram illustrating a user input voice analysis method according to an embodiment of the present invention.
5 is a flowchart illustrating a user input voice analysis method according to an embodiment of the present invention.
6 is a conceptual diagram illustrating a method of converting a user input voice into text according to an embodiment of the present invention.
7 is a conceptual diagram illustrating a user device according to an embodiment of the present invention.

It should be noted that the technical terms used in the present invention are only used to describe specific embodiments, and are not intended to limit the present invention. In addition, technical terms used in the present invention should be interpreted as meanings generally understood by a person having ordinary knowledge in the technical field to which the present invention belongs, unless defined otherwise. It should not be interpreted as a meaning or an excessively reduced meaning. In addition, when the technical term used in the present invention is a wrong technical term that does not accurately represent the spirit of the present invention, it should be understood as being replaced by a technical term that can be correctly understood by those skilled in the art. In addition, the general terms used in the present invention should be interpreted as defined in the dictionary or according to context before and after, and should not be interpreted as an excessively reduced meaning.

In addition, the singular expression used in the present invention includes a plural expression unless the context clearly indicates otherwise. In the present invention, terms such as “consisting of” or “comprising” should not be construed to include all of the various components or steps described in the present invention, and some of the components or some steps may not be included. It should be construed that it may or may further include additional components or steps.

Further, terms including ordinal numbers such as first and second used in the present invention may be used to describe elements, but the elements should not be limited by terms. The terms are used only to distinguish one component from another component. For example, the first component may be referred to as a second component without departing from the scope of the present invention, and similarly, the second component may be referred to as a first component.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings, but the same or similar elements will be given the same reference numbers regardless of the reference numerals, and redundant descriptions thereof will be omitted.

In addition, in the description of the present invention, when it is determined that detailed descriptions of related known technologies may obscure the subject matter of the present invention, detailed descriptions thereof will be omitted. In addition, it should be noted that the accompanying drawings are only for easy understanding of the spirit of the present invention and should not be interpreted as limiting the spirit of the present invention by the accompanying drawings.

Hereinafter, an embodiment of the present invention discloses a method for unlocking a user device based on speech recognition. For convenience of description, a method of unlocking a user device based on voice recognition is disclosed, but the voice recognition method according to an embodiment of the present invention not only unlocks a user device, but also other user authentication procedures required for the user device or It can also be used to control the operation of the device.

1 is a conceptual diagram illustrating a method for unlocking a user device based on speech recognition according to an embodiment of the present invention.

Hereinafter, in an embodiment of the present invention, a user device may be referred to in various terms. For example, a user equipment (UE) may be fixed or mobile, and a mobile station (MS), a mobile terminal (MT), a user terminal (UT), a subscriber station (SS), and a wireless device ), PDA (personal digital assistant), wireless modem (wireless modem), handheld device (handheld device), may be referred to in other terms such as wireless terminals.

As illustrated in FIG. 1, the user device 100 may be implemented to be used after releasing the lock state through authentication. The user device 100 may receive the user's voice signal 110 and perform user authentication. For example, the user may generate a voice signal 110 by pronouncing a specific word or sentence to unlock the user device 100. The user device 100 may receive the voice signal 110 and analyze the received voice signal 110 to determine whether to unlock the locked state 120. According to an embodiment of the present invention, the user device 100 compares the received voice signal 110 and the predetermined text to the user device 100 to release the locked state 120 of the user device 100 (130) You can decide whether or not to do it. The predetermined text may be a sentence or word set to release 130 the locked state 120 of the user device 100. Hereinafter, in an embodiment of the present invention, the predetermined text in order to release the locked state 120 of the user device 100 may be defined and used in the term reference text. In addition, the voice signal 110 input by the user to the user device 100 for unlocking 130 of the user device 100 may be defined and used in the term user input voice.

According to an embodiment of the present invention, it is determined whether the input voice signal 110 is human voice, and when the voice signal is human voice, the user input voice is changed to text, and the changed text and reference text are specified. Matching information can be calculated by matching units. A specific unit can be various units of a language such as a word, a syllable, a vocabulary, and a sentence. Hereinafter, in an embodiment of the present invention, a unit of a language performing matching is expressed as a token.

Matching information for determining whether to unlock the locked state 120 of the user device 100 may be the number of matching tokens and the matching rate. The user device 100 may determine whether to unlock 130 the locked state 120 of the user device 100 by comparing the calculated matching information with the lock setting release determination criteria.

The criterion for determining whether to unlock 130 the locked state 120 may be a value set by the user in advance, and may be a setting for the number of matching tokens and the matching rate between the user input voice and the reference text.

Hereinafter, in an embodiment of the present invention, a method for determining whether a user device 100 receives a user input voice and analyzes the received user input voice to determine whether to unlock the user device 100 is disclosed. do.

2 is a conceptual diagram illustrating a method of unlocking a user device 100 based on speech recognition according to an embodiment of the present invention.

As illustrated in FIG. 2, the user may pronounce the word “gil mobile phone” to release the locked state of the user device 100 and input it to the user device 100. The user device 100 may determine whether the input sound is a human voice based on the sound type analysis 200.

For example, the user device 100 determines whether the sound input through the sound type analysis 200 is a living sound, an inanimate sound, a human sound, an external noise, or a mechanical sound. Only when the input sound information can be analyzed. By using this method, it is possible to determine whether to unlock the user device 100 only for the user input voice while ignoring the sound other than the voice of the person input from the outside.

For example, the user device 100 may obtain characteristic information of the input sound for the sound type analysis 200 of the input sound. Human voice is an irregular signal with frequency and amplitude. Accordingly, when the characteristic information of the acquired sound is a certain pattern or exceeds a specific frequency wavelength, the user device 100 may determine that the input sound is not a human voice signal. When determining that the input sound is not a human voice signal, the user device 100 may not perform the additional user voice analysis 220. Conversely, when it is determined that the input signal is a human voice, an additional user voice analysis 220 may be performed to determine whether the user device 100 can be unlocked with the input voice.

As another example, the user device 100 may specifically determine whether the user's voice is correct by obtaining characteristic information of the user's voice based on the user's phone voice or a voice input by the user separately.

For example, when a user makes a phone call, the user's voice information may be acquired and the user's voice information may be learned to extract characteristics of the user's voice. The user device 100 may determine whether to unlock the lock by performing user voice analysis on only voices having the same characteristics as the voices of the user among voices input from the outside for unlocking.

When it is determined that the sound input to the user device 100 is the user input voice 210 based on the sound type analysis 200, the user voice analysis 220 may be performed as the next step.

The user speech analysis 220 may extract the similarity by changing the user input speech 210 to text and matching the changed text with the reference text 230 in token units. The user speech analysis 220 may use a speech-to-text conversion method, such as an automatic speech recognition (ASR) or text to speech (TST) method, to convert the user input speech 210 into text. .

For example, the user device 100 may convert a voice signal into text information of a'gil gil' based on the ASR when a sound of'gil gil' is input. For example, the user device 100 may analyze the user input voice in a specific unit such as a sentence level or a word level and convert it into text.

The user device 100 may determine whether to unlock the locked state of the user device 100 by comparing the text in which the user voice signal is converted and the reference text 230 stored in advance. The pre-stored reference text 230 is text previously stored in the user device 100 by the user, and may be text for determining whether to unlock the user device compared to the user input voice 210.

Also, the reference text 230 may be updated based on textualized information of a voice that can be unlocked among the user input voices 210.

According to an embodiment of the present invention, whether the user input voice 210 converted to text and the reference text 230 are similar enough to unlock the user device 100 may be determined according to user settings. A parameter that is a reference for determining whether the user input voice 210 converted into text and the reference text 230 are similar enough to unlock the user device may be referred to as a user setting parameter 240.

For example, the user setting parameter 240 may be the number of matching tokens, the matching rate, or the recognition rate between the user input voice converted into text and the reference text 230. When the user input voice converted into text and the reference text 230 are calculated to be equal to or greater than the threshold value of the user setting parameter 240 set in the user setting, the locked state of the user device 100 may be released. . Conversely, if it is less than the threshold of the user setting parameter 240, the locked state of the user device 100 may not be released.

For example, it may be assumed that the user setting parameter 240 is set to 1 number of matching tokens and the matching rate is 70%. In this case, whether the number of matching tokens is 1 or more and the matching rate is 70% or more by comparing the'Gil Dongi phone' recognized as text in the user device 100 and the reference text'I am Hong Gil Dong' I can judge. When the determination result satisfies the threshold of the set user setting parameter 240, the locking device of the user device 100 may be released. Conversely, when the text input by the user input voice 210 does not satisfy the set user setting parameter 240, the locking device of the user device 100 may not be released.

According to an embodiment of the present invention, the security level of the user device 100 may be determined according to the user setting parameter 240. That is, the threshold of the user setting parameters (token, matching rate, recognition rate, etc.) 240 may be adjusted to determine the security level of the user device 100. For example, the threshold of the user setting parameter 240 may be set to a high value to set the security degree high, and the threshold of the user setting parameter 240 may be set to a low value to set the security degree low. .

In addition, the user setting may be additionally set by selecting whether or not to'save the result' whether to update the reference text group based on the user input voice 210 in which the unlocking is performed. For example, in the case of a user input voice 210 that has unlocked the user device 100, such as a'kicker', it is added as a new reference text 230 that can unlock the user device 100 By doing so, the reference text group can be updated.

3 is a flowchart illustrating a method of determining a user setting parameter 240 according to an embodiment of the present invention and determining whether to release the locked state.

As shown in FIG. 3, first, a user setting parameter is determined (step S300).

The user setting parameter 240 is a parameter set to compare the voice information converted into text and the reference text 230 and may be a number of tokens, a matching rate, and the like. Depending on the security level of the user device 100, the user parameter 240 for unlocking the lock device may be adaptively set.

For example, if the number of tokens is set to 3 with the user setting parameter 240 and the matching rate is set to 50%, the number of matching tokens in the voice information converted into text and the reference text is 3 or more, and the matching rate is 50%. In the above case, the locked state of the user device 100 may be released. The minimum value of the user setting parameter 240 for releasing the locked state of the user device 100 may be referred to as a threshold of the user setting parameter 240.

Next, matching information is calculated based on the voice information converted into text and the reference text information (step S310).

The matching information is a value calculated by comparing voice information converted into text and reference text information. Matching information may be calculated to determine whether to unlock the locked state of the user device 100 by comparing the threshold value of the user setting parameter 240. When the user setting parameter 240 is the number of tokens and the matching rate, the matching information may also be the number of tokens and the matching rate.

Thereafter, it is determined whether the matching information satisfies the threshold of the user setting parameter (step S320).

As a result of matching information based on the user input voice converted into text and reference text information, it may be determined whether the threshold of the user setting parameter 240 is satisfied. For example, it may be determined whether the number of matching tokens calculated as a result of the matching is greater than the number of tokens being the user-set parameter 240, and the matching rate calculated as a result of the matching is greater than the matching rate of the user-set parameter 240. .

If the matching information satisfies the threshold of the user setting parameter 240, the locked state of the user device is released (step S330).

As a result of the determination, when the matching information satisfies the threshold of the user setting parameter 240, the locked state of the user device 100 may be released.

As described above, the user input voice converted to text may be updated to be used as reference text. Also, the input user voice may be used to calculate characteristic information of the user voice.

If the matching information does not satisfy the threshold value of the user setting parameter 240, the locking state of the user device 100 is not released and the determination step may end.

4 is a conceptual diagram illustrating a user input voice analysis method according to an embodiment of the present invention.

Referring to FIG. 4, a method for determining whether to unlock the locking device of the user device 100 through analysis of a user input voice when sound type analysis is determined to be human voice is disclosed.

In order to text the user input voice, the user device 100 may divide the user input voice into token units.

In addition, the user device 100 may provide a function of matching each token by comparing and storing the unique frequency wavelength of the vocabulary through individual voice pattern analysis. The last matched result can be stored in the cache.

For example, user voice feature information or average user voice feature information for individual syllables such as'road' and'dong' may be calculated from the user input voice. The user voice feature information is stored and may be used to determine whether the voice that is subsequently input is the user's voice or the voice of another person. In addition, it may be used for machine learning to textualize the voice input later.

For example, as illustrated in FIG. 4, the feature information 400 of the user's voice may be used to determine the user's input voice that is subsequently input as follows. When a user input voice of six letters's, mart, phone, open, open' is input, a letter outside the frequency range of the user's voice may be excluded based on the feature information 400 of the user's voice. If the syllable'o' is outside the frequency range of the user's voice, the user input voice may be recognized as'smartphone open' except for the syllable'o'. Through this method, the user input voice can be more accurately determined.

The voice feature information 400 of the user may be pitch information of the user voice as well as frequency information of the voice.

Hereinafter, the voice characteristic information of the user 400 discloses a method of calculating text from the user input voice based on the pitch information of the voice.

5 is a flowchart illustrating a method for analyzing a user input voice according to an embodiment of the present invention.

5 discloses a method of calculating text based on a user's voice for comparison with reference text.

Referring to FIG. 5, a pitch is detected in the user input voice (step S500).

The pitch in the user input voice can be detected in various ways. For example, the entire voice signal of the user input voice may be divided into short sections and analyzed. The analysis of the short section of the speech signal can be composed of a fixed analysis length of about 10 to 30 msec that can include a pitch period of one or more periods, assuming that the speech signal is a linear time-invariant system in a short period of time.

As another pitch detection method, the pitch can be detected based on the similarity of the voice signal. Since the voice signal maintains a quasi-periodic pattern in the voiced sound section, it may have a similar waveform of two adjacent pitch periods. Based on the quasi-periodic property, two adjacent sections can be selected to calculate a cross-correlation value, and the similarity between speech signals of two sections can be investigated. That is, when the length of two adjacent sections is a pitch period, the cross-correlation may be maximum. In addition, the pitch can be detected from the user's voice in various ways.

The feature vector is extracted from the user input speech (step S520).

Feature vectors may be extracted from the user input speech based on the pitch. The length of the analysis frame for extracting the feature vector from the user input speech may vary depending on the pitch. Accordingly, according to an embodiment of the present invention, the length of an analysis frame composed of each pitch period may be calculated, and a fast Fourier transform analysis size may be determined to extract a feature vector from the user input speech. Various methods can be used as a method of extracting a feature vector from a user input speech. The feature vector calculated from the user input voice may be used to extract only the user input voice from the learned and input sound information.

Based on the feature information of the user voice, a user input voice is extracted from sounds input to the user input device (step S540).

The feature information of the user voice may be information for determining whether it is a user voice, such as voice pitch information, feature vectors, frequency information, and the like. The user input voice to be generated as text may be extracted from the sound information input to the user device 100 based on the voice pitch information and the feature vector calculated based on the steps S500 and S520.

The extracted user input voice is converted into text (step S560).

Various methods can be used to convert the extracted user input speech into text. For convenience of description, an embodiment of the present invention discloses a method of converting an input voice into text based on word-level matching and sentence-level matching.

6 is a conceptual diagram illustrating a method of converting a user input voice into text according to an embodiment of the present invention.

As shown in FIG. 6, sound information input to the user device 100 may be analyzed as text after being pre-processed through the pre-processing step 600. For example, the pre-processing step 600 may be a process of extracting a user input voice for texting based on the above-described pitch information and feature vectors.

The user input voice after the pre-processing step 600 may be changed to the text 640 through word level or sentence level matching 630. In order to change to text 640 through word level and sentence level matching 630, an acoustic model 610 that models and compares the signal characteristics of speech and words or syllables corresponding to the recognized vocabulary A language model 620 that models linguistic order relationships may be used.

The acoustic model 610 may use a direct comparison method of setting a recognition object as a feature vector model and comparing it with a feature vector of an input signal or a statistical method of statistically processing the feature vector of the recognition object.

The direct comparison method is a method of setting units of words, phonemes, and the like to be recognized as a feature vector model and comparing how similar the input speech is. A vector quantization method may be used as one of the direct comparison methods. The vector quantization method is a method of matching the feature vectors or coefficients obtained above with a codebook, which is an existing model, to encode the representative values, and comparing the coded values with each other.

The statistical model method is a method of constructing a unit for a recognition object into a state sequence and using the relationship between the state columns. The status column can display complex models, so most speech recognition uses this technique except simple word recognition. The method of using the relationship between the status columns is a method of using a temporal array relationship such as Dynamic Time Warping (DTW), a statistical method of comparing probability values, means, and variances such as the hidden markov model (HMM), and artificial neural networks. There is a way to use it.

DTW is a method of compensating for a difference in the time axis when compared with the existing model considering the dynamic characteristics of the voice whose signal length varies with time even if the same person pronounces the same.

HMM estimates the state transition probability and the observation probability of the output symbol through the training data after assuming that the speech is a Markov process having the state transition probability and the observation probability of the output symbol in each state, and the voice input from the estimated model It is a recognition method that calculates the probability of occurrence.

On the other hand, the language model 620 for modeling linguistic order relationships such as words or syllables reduces acoustic ambiguity and reduces errors in recognition by applying the order relationships between units constituting languages to units obtained in speech recognition. have. In general, in the case of continuous speech recognition targeting a large vocabulary, it is important to combine linguistic knowledge in the form of a language model 620 because it relies heavily on linguistic knowledge inherent in input speech.

The language model 620 includes a statistical language model and a model based on a Finite State Automata (FSA). In the statistical language model, specifically, chain probabilities of words such as Unigram, Bigram, and Trigram can be used. The case based on the finite state network is mainly used when the recognition target area is not large, and the type of the spoken character is patterned. It can be used to control robots or to perform application programs or simple commands on a desktop on a computer.

According to an embodiment of the present invention, the extracted user input voice may be calculated as text 640 after performing word level matching and sentence level matching 630 based on an acoustic model or a language model. The calculated text 640 may be compared with the reference text as described above to determine whether to unlock the user device through token matching according to a user setting.

7 is a conceptual diagram illustrating a user device according to an embodiment of the present invention.

Referring to FIG. 7, the user device may include a sound process unit (SPU).

The SPU included in the user device 100 may include a sound type analysis unit 710, a vocabulary analysis unit 720, an individual voice pattern analysis unit 730, a user setting unit 740, and a processor 750. .

The sound type analysis unit 710 may determine whether the input sound is a human voice. For example, the sound type analysis unit 710 may determine whether the input sound is a living sound or an inanimate sound, and in the case of a living body, whether it is a human or an animal. The sound type analyzer 710 may determine whether the input sound is a human voice based on the characteristics (frequency, amplitude, etc.) of the input sound. The sound type analyzer 710 may further determine whether the input voice is more specifically a user's voice.

For example, characteristics of user voice information may be extracted based on voice previously input by the user or voice when the user makes a phone call. The characteristics of the extracted user voice information may be used by the sound type analysis unit 710 to determine whether it is the user's voice.

The lexical analysis unit 720 may classify the input user's voice into a specific unit. For example, the vocabulary analysis unit 720 may separate the user voice input in units of tokens.

The voice pattern analysis unit 730 for each individual may change the voice to text, and match the changed text for each token unit (eg, syllables and words) to extract similarity. The user voice may be changed to text based on the techniques described above in FIGS. 5 and 6. The voice pattern analysis unit 730 for each individual may change only voices within an effective range into texts in consideration of characteristics of the user's voice when the voice is changed to text. For example, if the user's voice characteristic is out of the input voice in consideration of the voice characteristic of the previously input user, the corresponding information may be excluded.

In addition, the individual voice pattern analysis unit 730 may compare the user voice signal analyzed with text and the reference text based on the user setting parameter 240 set in the user setting unit. It may be determined whether the analyzed result satisfies the user setting parameter 240. For example, the user setting parameter may be the same number of tokens, matching rate, recognition rate, etc. between the reference text and the user's voice changed to text.

The user setting unit 740 may compare a user voice converted to text with a reference text and determine a user setting parameter for unlocking the user device with the input voice. For example, the user setting parameter may be a number of tokens, a matching rate, and a recognition rate for comparing the reference text and the user's voice changed to text.

The processor 750 may be implemented to control operations of the sound type analysis unit 710, the vocabulary analysis unit 730720, the individual voice pattern analysis unit 750730, and the user setting unit 740. In addition, the processor 750 may be implemented to release the locked state of the user device when the input sound signal can release the locked state of the user device as a result of analysis of the individual voice pattern analysis unit. Conversely, the processor 750 adds the sound signal input as a result of analysis by the individual voice pattern analysis unit 750730 without unlocking the user device 100 when the input sound signal cannot be unlocked. It can be implemented to command the user to input the voice once again.

By using the method of recognizing the human voice, which is the biometric technology, it is possible to provide a more convenient user-friendly environment such as existing key input and to increase security. It also helps children and people with disabilities by using a voice-based interface. In addition, the user can determine the recognition level according to the setting value of the user setting parameter 240 and use it flexibly and according to the situation.

The voice recognition method according to an embodiment of the present invention can be used not only for unlocking various devices, but also for voice recognition based personal authentication and information input in services other than the lock device.

A method for unlocking a user device based on voice according to an embodiment of the present invention can be written in a computer program, and codes and code segments constituting the computer program can be easily deduced by a computer programmer in the art. In addition, the computer program is stored in a computer-readable information storage medium (computer readable media), and is read and executed by a computer or a user device according to an embodiment of the present invention to implement a method for unlocking a user device. have.

The information storage medium includes a magnetic recording medium, an optical recording medium and a carrier wave medium. A computer program that implements a method for releasing a locked state of a user device according to an embodiment of the present invention may be stored and installed in an internal memory such as a user device. Alternatively, an external memory such as a smart card that stores and installs a computer program that implements a method for releasing a locked state of a user device according to an embodiment of the present invention may be mounted on a terminal or the like through an interface.

The above-described contents may be modified and modified without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the claims below, and all technical spirits within the scope equivalent thereto should be interpreted as being included in the scope of the present invention.

The present invention determines whether the sound information input to the user device is human voice, and when the sound information is human voice, generates converted text based on the sound information and compares the converted text with the reference text to lock the user device You can decide whether to release it. This method can be utilized in a user device to provide a more convenient user-friendly environment such as existing key input and to increase the security of the user device. In addition, since the present invention uses a voice-based interface for unlocking a user device, it can be applied to a user device used by children and people with disabilities. Furthermore, the recognition level may be applied to the user device to adaptively set whether or not the user device is unlocked by determining the recognition level according to the setting value of the user setting parameter.

100: user device 110: voice signal
120: locked state 130: unlocked state
200: sound type analysis 210: user input voice
220: user voice analysis 230: reference text
240: User setting parameter 400: User voice feature information
600: pre-processing step 610: acoustic model
620: Language model 630: Word level matching and sentence level matching
640: text 700: SPU
710: sound type analysis unit 720: vocabulary analysis unit
730: personal voice pattern analysis unit 740: user setting unit
750: processor

Claims (11)

A sound type analyzer configured to determine whether sound information input to the user device is a human voice;
When the sound information is the voice of the person, reference text that is text information that is a criterion for determining whether to generate a converted text based on the sound information and determine whether to unlock the locked device of the user device with the converted text A voice pattern analysis unit that is implemented to determine whether to unlock the user device based on the matching result calculated by comparing and a preset user setting parameter; And
A processor implemented to unlock the user device when the matching result satisfies the threshold of the user setting parameter
It includes,
The sound type analysis unit determines that the voice information is a human voice when there is no pattern in the sound information and the human voice frequency band, and if the voice information is not a human voice, the sound information is ignored to unlock the user device. And
The matching result includes information on the number of tokens matched between the converted text and the reference text and information on a matching rate,
The user setting parameter is threshold information for the number of tokens set to determine whether to unlock the user device and threshold information for the matching rate. User device. delete According to claim 1,
The number of matching tokens is calculated based on the number of words mapped by performing mapping for each word included in the converted text and the reference text,
The matching rate is calculated based on the number of syllables that match in the mapped word, and the user device performs unlocking based on speech. delete According to claim 1,
The individual voice pattern analysis unit divides the sound information into token units, pre-processes the sound information divided into token units based on feature information of the user's voice, and converts the pre-processed sound information into the text to convert text Is implemented to generate
A user device for unlocking based on voice, characterized in that the feature information of the user voice is frequency information. delete delete Determining whether sound information input to a user device is a human voice;
If the sound information is the voice of the person, generating converted text based on the sound information;
Whether to unlock the user device based on the matching result calculated by comparing the converted text and the reference text, which is text information that is a judgment criterion for determining whether to unlock the user device, and a user setting parameter. Determining; And
And when the matching result satisfies the threshold of the user setting parameter, unlocking the user device.
The step of determining whether the voice is the human voice is determined by the voice of the human when the pattern does not exist in the voice information and the voice frequency band of the human voice is ignored. Further comprising the step of preventing the user device from being unlocked,
The matching result includes information on the number of tokens matched between the converted text and the reference text and information on a matching rate,
The user setting parameter is threshold information for the number of tokens set to determine whether or not to unlock the user device, and threshold information for the matching rate. delete delete delete

Images