KR101854369B1 - Apparatus for providing speech recognition service, and speech recognition method for improving pronunciation error detection thereof - Google Patents

Abstract

The present invention distinguishes and stores an erroneous pronunciation dictionary that defines a reference pronunciation for each word and an error dictionary for error detection that defines a possible erroneous pronunciation type for each word, The speech recognition unit recognizes a word corresponding to the user voice input based on the voice recognition pronunciation dictionary and extracts an error pronunciation for the recognized word from the error detection pronunciation dictionary, A service device for detecting an error of a user's pronunciation and providing the detected error to a terminal device; And a terminal device for receiving the user's voice from the user, requesting the error detection by transmitting the user's voice to the service device, receiving the error detection result for the pronunciation of the user's voice input from the service device, The present invention provides a speech recognition system for improving the pronunciation detection ability by detecting an error in a state of knowing an error type, thereby performing fast and accurate speech recognition and detecting an error in a spoken word.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a speech recognition service,

The present invention relates to speech recognition for performing speech recognition on user speech, and more particularly to an apparatus for providing speech recognition services for improving the detection ability of a foreign language pronunciation error caused by mother language interference, And a speech recognition method for improving the ability.

In general, speech recognition uses an acoustic model to measure human voice patterns and similarities. In this case, phoneme is often used as the minimum unit of the acoustic model. In addition, a dictionary in which phonetic column information that can be spoken per word is recorded is called a lexicon. One word may also be spoken as several phonemes, so a plurality of information corresponding to different phonemes are added to words that are spoken by a plurality of phonemes.

In conventional speech recognition, only the pronunciation dictionary for correct pronunciation (standard pronunciation) is used to recognize a word corresponding to the user's utterance.

In the conventional speech recognition technology, it is difficult to correctly recognize the pronunciation errors due to the native language interference of the foreign language learners. In particular, it is impossible to recognize the pronunciation errors due to the standard pronunciation and the native language interference.

In order to accurately recognize words even in pronunciation errors due to native language interference, it is necessary to add more phonemes in the phonetic dictionary. If too many phonemes in one word are used, The recognition performance deteriorates.

The present invention is characterized in that a pronunciation dictionary used for speech recognition is divided into two parts for speech recognition and error detection, and speech recognition is performed using a speech recognition pronunciation dictionary to find correct words and a pronunciation dictionary for error detection is applied An apparatus for providing a speech recognition service capable of improving speech recognition performance and error detection performance by detecting a pronunciation error existing in a word uttered by a speaker, and a speech recognition method for improving the error pronunciation detection capability thereof .

As a means for solving the above-mentioned problems, the present invention provides a communication system comprising: a communication unit for transmitting and receiving data through a communication network; A phonetic phonetic dictionary for defining a phoneme sequence that can be generated for each word for speech recognition, and a phonetic dictionary for error detection that defines an error phonological column that can be generated for each word to detect errors in user utterance ; And a control unit for recognizing a word corresponding to the user's voice based on the voice recognition pronunciation dictionary when the user's voice is received from the terminal device through the communication unit, And a service provider for detecting a pronunciation error and providing a result of speech recognition and an error detection result to the terminal device.

In the service apparatus according to the present invention, the phonetic dictionary for voice recognition and the phonetic dictionary for error detection include error phonemes that can be generated by mother language interference, and the phonetic phonetic dictionary for voice recognition has a smaller phoneme And heat.

In the service apparatus according to the present invention, the service providing unit may include an error detection module that detects an error type including at least one of pronunciation, longitude, intonation, and accent with respect to the user's voice.

In the service apparatus according to the present invention, the error-detection pronunciation dictionary further includes an error occurrence frequency rate for each error phoneme string, and the service providing unit stores the error sound sequence and the user speech in the order of high error frequency rate The error can be detected by comparing.

In the service apparatus according to the present invention, when providing the error detection result to the terminal apparatus, the service providing unit may further provide information on the correction method or error cause for each error type.

The service provider of the service apparatus according to the present invention may be configured such that when the user's voice is not matched with the standard phoneme string but the similarity is within a predetermined range and the phoneme string corresponding to the phonetic dictionary for error detection does not exist, And an error management module for adding a corresponding phoneme string to the phonetic dictionary for error detection.

Another object of the present invention is to provide a speech recognition dictionary for speech recognition which defines a reference phoneme sequence that can be generated for each word for speech recognition and a speech recognition dictionary for speech recognition, A storage unit for storing a phonetic dictionary for error detection that defines possible error phonemes separately; An input unit for receiving a user request; An audio processing unit for receiving a user's voice; According to a request of a user input through an input unit, a word corresponding to a user voice input through the audio processing unit is recognized using a voice recognition pronunciation dictionary, and a pronunciation error And outputs a voice recognition result and an error detection result; And an output unit for outputting a speech recognition result and an error detection result.

In the terminal device according to the present invention, the phonetic dictionary for speech recognition and the phonetic dictionary for error detection include error phonemes that can be generated by the mother language interference, and the phonetic phonetic dictionary for speech recognition has a smaller phoneme And heat.

In the terminal device according to the present invention, the control unit can further detect an error type including at least one of pronunciation, long end, intonation, and accent with respect to the user's voice.

In the terminal device according to the present invention, the phonetic dictionary for error detection further includes an error occurrence rate for each erroneous phoneme string, and the controller compares the error phoneme string and the user voice in the order of high error frequency rate So that an error can be detected.

In the terminal device according to the present invention, the control unit may further provide information on a correction method or an error cause for each error type when providing the error detection result.

In the terminal device according to the present invention, the control unit can receive and store the pronunciation dictionary for voice recognition and the pronunciation dictionary for error detection from the service apparatus through the communication unit.

The present invention also provides a method of recognizing a word corresponding to a user's voice by using a phonetic pronunciation dictionary that defines a reference phoneme string that can be generated for each word for speech recognition, ; And detecting a pronunciation error included in the user's voice by using a pronunciation dictionary for error detection that defines an error phoneme sequence that can be generated for each word in order to detect an error in user utterance A speech recognition method for improving the detection capability is provided.

In the speech recognition method according to the present invention, the phonetic dictionary for speech recognition and the phonetic dictionary for error detection include error phonemes that can be generated by mother language interference, Phonemes and phonemes.

The speech recognition method according to the present invention may further include the step of detecting an error type including at least one of pronunciation, long end, intonation, and stress with respect to the detected error.

The present invention relates to a method of constructing a phonetic dictionary that defines phonemes that can be generated for each word for speech recognition. The phonetic dictionary is divided into speech recognition and error detection, and a word corresponding to the user speech The speech recognition and the error detection performance can be improved by detecting a pronunciation error included in the user's voice through the pronunciation dictionary for error detection.

In particular, the present invention is intended to include an error phoneme sequence that can be generated by a user's mother language interference in the construction of a phonetic pronunciation dictionary and a phonetic dictionary for error detection, By reducing the number of phonemes, it is possible to improve speech recognition performance by improving speech recognition speed and accuracy by learning with the least compressed speech sequence in speech recognition and by applying all possible error phonemes in error detection have.

FIG. 1 is a block diagram of a speech recognition system for improving error detection capability according to the present invention.
FIG. 2 is a block diagram illustrating a configuration of a service apparatus in a speech recognition system for improving error detection capability according to the present invention.
3 is a flowchart illustrating a speech recognition method for improving error pronunciation detection capability performed through the service apparatus of the present invention.
FIG. 4 is a block diagram illustrating a configuration of a terminal device for providing a speech recognition service for improving error detection capability according to another embodiment of the present invention. Referring to FIG.
FIG. 5 is a flowchart illustrating a speech recognition method for improving the error pronunciation detection capability according to another embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, detailed description of well-known functions or constructions that may obscure the subject matter of the present invention will be omitted. It should be noted that the same constituent elements are denoted by the same reference numerals as possible throughout the drawings.

The terms and words used in the present specification and claims should not be construed in an ordinary or dictionary sense, and the inventor shall properly define the terms of his invention in the best way possible It should be construed as meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention, and not all of the technical ideas of the present invention are described. Therefore, It is to be understood that equivalents and modifications are possible.

FIG. 1 is a block diagram of a speech recognition system for improving error detection capability according to the present invention. Referring to FIG. 1, a speech recognition system according to an embodiment of the present invention includes a terminal device 100, a service device 200 , And a network 300. [

In one embodiment of the present invention, the speech recognition service may be based on a server-based computing approach. Here, the service-based computing method refers to a method in which a method of providing a voice recognition service for improving the error pronunciation detection capability according to the present invention is performed in an arbitrary device connected via a network, and a terminal device performs only input and output. Hereinafter, for convenience of description, an apparatus for providing a speech recognition service for improving the error pronunciation detection capability according to the present invention will be described as a service apparatus 200.

The service apparatus 200 provides a speech recognition service for improving the error pronunciation detection capability according to the present invention to the terminal device 100 via the network 300. [ More specifically, the service device 200 separately stores an pronunciation pronunciation dictionary in which a reference phoneme string for each word is defined, and a pronunciation dictionary for error detection that defines an error phoneme sequence that can be generated for each word, When an error detection for the user's voice input is requested from the speech recognition apparatus 100, a word corresponding to the user's speech input is recognized based on the pronunciation dictionary for speech recognition, and an error pronunciation for the recognized word is extracted And provides it to the terminal device 100.

The phonetic dictionary for speech recognition and the phonetic dictionary for error detection may include an error phoneme sequence that can be generated by the mother language interference. However, the phonetic dictionary for speech recognition has a smaller phonetic sequence than the phonetic dictionary for error detection . Accordingly, the present invention can improve not only speech recognition performance but also error detection performance.

The service device 200 may operate in a server-client computing manner or may operate in a cloud computing basis. That is, it is possible to provide at least one of the computer resources, for example, hardware and software, required for proceeding with the speech recognition service for improving the error pronunciation detection capability, to the service apparatus 200.

The terminal device 100 may be any of various types of devices used by a user and may be a personal computer (PC), a notebook computer, a mobile phone, a tablet PC, a navigation terminal, a smart phone, A PDA (Personal Digital Assistants), a Smart TV, a PMP (Portable Multimedia Player), and a digital broadcast receiver. Of course, this is merely an example, and it should be construed as a concept including a device that is currently developed, commercialized, or capable of all communication to be developed in the future, in addition to the above-described examples.

Such a terminal device 100 can be used by a user who requests a speech recognition service for improving the false pronunciation detection capability. In the speech recognition system for improving the error sound detection capability according to the present invention, the terminal device 100 receives user's voice from a user, transmits the user's voice to the service device 200 to request error detection, Receives speech recognition and error detection results for the user's voice, and outputs the result to the user.

The network 300 provides a path for transmitting and receiving data between the service device 200 and a plurality of terminal devices 100. [ The network 300 is an i-bimetry that provides a large capacity data transmission / reception service and a data service without a disruption phenomenon through an internet protocol (IP). The i-bimet network 300 is an i-bimet network IP network. Also, the network 300 includes a 3G mobile network including a wired network, a Wibro (Wireless Broadband) network, a WCDMA, a 3.5G mobile network including an HSDPA (High Speed Downlink Packet Access) , A fourth generation mobile network, a satellite network, and a wireless LAN including a Wi-Fi network.

In the present invention, the terminal device 100 performs only an output function for providing a voice recognition service for improving the false pronunciation detection capability, and thus the service device 200 will be mainly described below.

2 is a block diagram illustrating a configuration of a service apparatus 200 for providing a speech recognition service for improving error detection capability according to an embodiment of the present invention. In FIG. 2, the configuration of the service apparatus 200 is expressed as a functional unit. However, in actual implementation, the service apparatus 200 may be distributed to a plurality of server apparatuses, or may be implemented in one server apparatus.

Referring to FIG. 2, in the speech recognition system for improving the error sound detection capability of the present invention, the service apparatus 200 may include a communication unit 210, a storage unit 220, and a service providing unit 230 .

The communication unit 210 exchanges data with the terminal device 100 through the network 300.

The storage unit 220 is a means for storing data and programs for operation of the service apparatus 200. In particular, in order to provide a speech recognition service for improving the error pronunciation detection capability according to the present invention, A phonetic pronunciation dictionary 221 in which columns are defined, and a phonetic dictionary 222 for error detection in which error phonemes are generated for each word are separately stored. In addition, the error detection pronunciation dictionary 222 further includes an occurrence frequency rate for each error phoneme sequence, so that the error type can be detected by comparing the error type with the input voice in the order of the occurrence frequency rate at the time of error detection.

The service providing unit 230 recognizes a word corresponding to the user's voice input on the basis of the voice recognition pronunciation dictionary 221 when the error detection of the user voice input is requested from the terminal device 100, A possible erroneous phoneme string for a recognized word is extracted from the phonetic pronunciation dictionary 222, and the erroneous phoneme string is compared with a user's voice to detect an error with respect to user pronunciation and provide the detected error to the terminal device 100.

In addition, in the case where the pronunciation is not equal to the standard pronunciation but corresponds to a predetermined similarity degree range and does not exist in the pronunciation dictionary for error detection 222, A phoneme string corresponding to speech can be added to the pronunciation dictionary 222 for error detection and stored.

The service providing unit 230 includes a voice recognition module 231, an error detection module 232, and an error management module 233 for providing a voice recognition service for improving the error sound detection capability according to the present invention .

The voice recognition module 231 recognizes a word corresponding to the user voice input based on the voice recognition pronunciation dictionary 221.

The error detection module 232 extracts a plurality of error phonemes that can be generated for the recognized word from the phonetic dictionary for error detection, and compares the generated error phonemes with user speech to detect errors included in the user's voice. The error phoneme string can be detected in comparison with the user speech in the order of the highest error frequency rate.

When the pronunciation corresponding to the user voice corresponds to the standard pronunciation and the predetermined similarity degree range and corresponds to a new error phoneme string that does not exist in the pronunciation dictionary for error detection 222, 222). The addition of the new erroneous phoneme string may be made at the request of the user.

The voice recognition module 231, the error detection module 232 and the error management module 233 may be realized by software or hardware or a combination of software and hardware. For example, the voice recognition module 231, the error detection module 232, 220 and executed by the service providing unit 230. [0050]

FIG. 3 is a flowchart illustrating a method for providing a speech recognition service for improving error detection capability according to the present invention.

3, the service apparatus 200 includes an pronunciation pronunciation dictionary 221 in which a reference phoneme string for each word is defined, and an error detection pronunciation dictionary 222 that defines an error phoneme string that can be generated for each word. (S105).

If a voice is input by the user to the terminal device 100 in step S110, the terminal device 100 may transmit the user voice to the service device 200 to request voice recognition and error detection in step S115. Accordingly, the service apparatus 200 recognizes a word corresponding to the received user voice on the basis of the pronunciation dictionary for voice recognition (S120).

Then, the error phonetic string corresponding to the word is compared with the user voice in the pronunciation dictionary for error detection, and an error existing in the pronunciation of the user is detected (S125). At this time, the service device 200 may further detect an error type including at least one of a pronunciation error, a long end error, an intonation error, and an accent error in relation to an error detected from the user's voice. Also, in comparing a plurality of error phonemes that can be generated for a recognized word with a user's voice, the occurrence frequency rate is stored for each error phoneme string, and the error phoneme string is compared with the user speech , The error can be detected more quickly.

The service device 200 provides the voice recognition and error detection result to the terminal device 100 in step S130 and the terminal device 100 transmits the voice recognition and error detection result received from the service device 200 to the user (S135).

At this time, when providing the error detection result, the service apparatus 200 can further provide a correction method or an error cause for the detected error type.

On the other hand, when the user's voice corresponds to a new error type that does not exist in the pronunciation dictionary for error detection and corresponds to the standard pronunciation and predetermined similarity degree range (S140), the service device 200 displays the pronunciation dictionary for error detection 222 The phonetic dictionary for error detection 222 can be updated by adding a phoneme string corresponding to the user's voice as a new erroneous phoneme sequence in step S145. At this time, the final selection of the error phoneme string to be added can be made by the user.

In another embodiment of the present invention, the provision of the speech recognition service for improving the error pronunciation detection capability may be performed based on the terminal device 100. [

FIG. 4 is a block diagram illustrating a configuration of a terminal device 100 for providing a voice recognition service for improving error detection capability according to another embodiment of the present invention.

4, a terminal device 100 according to the present invention includes an input unit 110, an output unit 120, an audio processing unit 130, a storage unit 140, and a control unit 150 .

The input unit 110 is a means for generating a user input signal for controlling or operating the terminal device 100 according to a user's operation, and may be implemented by various types of input means. For example, the input unit 110 may include at least one of a key input unit, a touch input unit, a gesture input unit, and a voice input unit. The key input means generates a signal corresponding to the key according to the key operation, and corresponds to a keypad and a keyboard. The touch input means is a touch pad, a touch screen, and a touch sensor, which recognize an input operation by sensing a user's operation of touching a specific portion. The gesture input means recognizes a specific operation, such as a shaking or moving operation of the terminal device, an approach to the terminal device, a blinking operation, etc., as a specific input signal, such as a geomagnetic sensor, A sensor, a camera, an altimeter, a gyro sensor, and a proximity sensor.

The output unit 120 is an output means for outputting an interface screen between the terminal device 100 and the user, and displays speech recognition and error detection results. For example, the output unit 120 may be a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT), a light emitting diode (LED), an organic light emitting diode (OLED) Matrix Organic Light Emitting Diodes), a flexible display, and a three-dimensional display.

The audio processing unit 130 is a means for processing input and output of voice and sound, and receives a user voice for voice recognition. The audio processing unit 130 may include a microphone capable of inputting audio and a speaker capable of outputting audio.

The storage unit 140 is a means for storing data or programs necessary for operation of the terminal apparatus 100 and can basically store an operating program (OS) of the terminal apparatus 100 and one or more application programs. In addition, in the present invention, the storage unit 140 includes an edible pronunciation dictionary 141 in which a reference phoneme string for each word is defined to provide a speech recognition service for improving the error pronunciation detection capability, And a phonetic dictionary 142 for error detection that defines possible error phonemes are separately stored. The storage unit 140 may be any type of storage such as a RAM (Read Only Memory), a ROM (Read Only Memory), a hard disk (HDD), a flash memory, a CD- Media.

The control unit 150 controls the entire operation of the terminal device 100 and basically operates based on the operating program stored in the storage unit 150 to establish a basic platform environment of the terminal device 100, Depending on the selection, the application is executed to provide arbitrary functions. In another embodiment of the present invention, when the voice recognition and the error detection are requested by the user through the input unit 110, the control unit 150 controls the audio processing unit 130 based on the voice recognition pronunciation dictionary 141 After recognizing a word corresponding to the input user voice, an error phoneme sequence for the recognized word is extracted from the error-detecting pronunciation dictionary 142, and the error phoneme sequence is compared with the user's voice, And outputs the detection result. The control unit 150 may include a voice recognition module 151, an error detection module 152, and an error management module 153.

The voice recognition module 151 recognizes a word corresponding to the user voice based on the voice recognition pronunciation dictionary 141.

The error detection module 152 extracts an error phoneme string including an error for at least one of pronunciation, long end, intonation, and accent with respect to the recognized word from the phonetic dictionary for error detection, and compares the extracted error phoneme string with the user voice And detects a pronunciation error included in the user voice. At this time, the occurrence frequency rate is stored for each erroneous phoneme string, and the error phoneme string can be compared with the user speech in order of the occurrence frequency ratio.

If the pronunciation of the inputted user voice corresponds to a new error type that does not exist in the error detection pronunciation dictionary 142 and the standard pronunciation and the predetermined similarity degree range, the error management module 153 sets the phoneme string of the user voice A new error type is added to the error detection pronunciation dictionary 142. [

The voice recognition module 151, the error detection module 152 and the error management module 153 may be implemented by software or hardware or a combination of software and hardware. For example, the voice recognition module 151, the error detection module 152, 140, and executed by the control unit 150. [0033] FIG.

FIG. 5 is a flowchart illustrating a method of providing a speech recognition service for improving error detection capability according to another embodiment of the present invention.

5, the terminal device 100 includes an edible pronunciation dictionary 141 in which a reference phoneme string for each word is defined, and an error detection pronunciation dictionary 142 that defines an error phoneme string that can be generated for each word, (S205). The voice recognition pronunciation dictionary 141 and the error detection pronunciation dictionary 142 can be received from the service device 200 and acquired.

When a user voice is input (S210), a word corresponding to the user voice is recognized based on the voice recognition pronunciation dictionary 141 (S215). Subsequently, a plurality of error phonemes including the possible pronunciation errors, long and short errors, intonation errors, and accent errors for the recognized words are extracted from the pronunciation dictionary 142 for error detection, and compared with the user speech, And detects the included erroneous pronunciation (S220). At this time, corresponding to the detected error pronunciation, an error type including at least one of a pronunciation error, a long end error, an accent error, and an accent error can be further detected. In addition, when an error is detected, the errors of the speech input words can be detected more quickly by comparing the pronunciation of the user's speech with the error phonological sequence in the order of the occurrence frequency ratio.

The terminal device 100 outputs speech recognition and error detection results through the output unit 120 and provides the result to the user (S225). At this time, the error detection result may further assist in the learning of a foreign language by further including the error type of each word or the cause of error in the pronunciation correction method or pronunciation of the detected error type.

On the other hand, if the voice input is a new error type that does not exist in the pronunciation dictionary for error detection (S230), the terminal device 100 displays the pronunciation dictionary for error detection 142 The pronunciation dictionary 142 for error detection may be updated by adding a new error type (S235).

The speech recognition method for improving the error sound detection capability according to the present invention can be implemented in software form readable by various computer means and recorded in a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, and the like, alone or in combination. Program instructions to be recorded on a recording medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. For example, the recording medium may be an optical recording medium such as a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, a compact disk read only memory (CD-ROM), a digital video disk (DVD) Includes a hardware device that is specially configured to store and execute program instructions such as a magneto-optical medium such as a floppy disk and a ROM, a random access memory (RAM), a flash memory, do. Examples of program instructions may include machine language code such as those generated by a compiler, as well as high-level language code that may be executed by a computer using an interpreter or the like. Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be apparent to those skilled in the art. Furthermore, although specific terms are used in this specification and the drawings, they are used in a generic sense only to facilitate the description of the invention and to facilitate understanding of the invention, and are not intended to limit the scope of the invention.

The speech recognizing dictionary and the error detecting dictionary are separately provided, so that the speech recognizing is performed through the speech recognizing dictionary to find the correct word, and the error detecting dictionary including the error type for the word is used to present And detects an error in a state in which the error type is known, so that it is possible to perform fast and accurate speech recognition and to detect an error in a spoken word.

100: terminal device 110: input unit 120: output unit
130: Audio processing unit 140:
141: Speech recognition pronunciation dictionary 142: Error detection pronunciation dictionary
150: control unit 151: voice recognition module
152: Error detection module 153: Error management module
200: service apparatus 210: communication unit 220:
221: Speech recognition pronunciation dictionary 222: Error detection speech dictionary
230: Service provisioner 231: Voice recognition module
232: error detection module 233: error management module 300: network

Claims (15)

A communication unit for transmitting and receiving data through a communication network;
A phonetic phonetic dictionary for defining a phoneme sequence that can be generated for each word for speech recognition, and a phonetic dictionary for error detection that defines an error phonological column that can be generated for each word to detect errors in user utterance ; And
When a user's voice is received from the terminal device through the communication unit, recognizes a word corresponding to the user's voice on the basis of the pronunciation dictionary for voice recognition, and then uses the pronunciation dictionary for error detection to pronounce A service providing unit that detects an error and provides a voice recognition result and an error detection result to the terminal device;
Lt; / RTI >
The service providing unit,
When the user's voice is not matched with the standard phoneme string but the degree of similarity is in a predetermined range and the phoneme string corresponding to the phonetic dictionary for error detection does not exist, And an error management module for adding the voice recognition service to the voice recognition service. The method according to claim 1,
Wherein the phonetic dictionary for speech recognition and the phonetic dictionary for error detection include error phonemes that can be generated by mother language interference and the phonetic pronunciation dictionary includes fewer phonemes than the phonetic dictionary for error detection A service device providing a voice recognition service. The method of claim 1, wherein the service providing unit
And an error detection module for detecting an error type including at least one of pronunciation, longitude, intonation, and accentuation with respect to the user voice. The method according to claim 1,
Wherein the error detection pronunciation dictionary further includes an error occurrence frequency rate for each error phoneme string,
Wherein the service providing unit detects an error by comparing the error phoneme string and the user voice in the order of high error frequency rate when an error is detected. The method of claim 1, wherein the service providing unit
Wherein when the error detection result is provided to the terminal apparatus, the terminal apparatus further provides information on a correction method or an error cause for each error type. delete A phonetic phonetic dictionary for defining a phoneme sequence that can be generated for each word for speech recognition, and a phonetic dictionary for error detection that defines an error phonological column that can be generated for each word to detect errors in user utterance ;
An input unit for receiving a user request;
An audio processing unit for receiving a user's voice;
A speech recognition apparatus for recognizing a word corresponding to a user's voice input through the audio processing unit using the speech recognition dictionary for speech recognition in response to a user's request input through an input unit, And outputting the speech recognition result and the error detection result; And
An output unit for outputting the speech recognition result and the error detection result;
Lt; / RTI >
Wherein,
When the user's voice is not matched with the standard phoneme string but the degree of similarity is in a predetermined range and the phoneme string corresponding to the phonetic dictionary for error detection does not exist, And the voice recognition service is added to the voice recognition service. [8] has been abandoned due to the registration fee. 8. The method of claim 7,
Wherein the phonetic dictionary for speech recognition and the phonetic dictionary for error detection include error phonemes that can be generated by mother language interference and the phonetic pronunciation dictionary includes fewer phonemes than the phonetic dictionary for error detection A terminal device providing a voice recognition service. [Claim 9 is abandoned upon payment of registration fee.] 8. The apparatus of claim 7, wherein the control unit
Wherein the error detection unit further detects an error type including at least one of pronunciation, longitude, intonation, and accent with respect to the user voice. [Claim 10 is abandoned upon payment of the registration fee.] 8. The method of claim 7,
Wherein the error detection pronunciation dictionary further includes an error occurrence frequency rate for each error phoneme string,
Wherein the controller detects an error by comparing the error phoneme string with the user speech in the order of the error frequency rate when the error is detected. [Claim 11 is abandoned upon payment of the registration fee.] 8. The apparatus of claim 7, wherein the control unit
Wherein the information providing unit further provides information on a correction method or an error cause for each error type when providing the error detection result. 8. The apparatus of claim 7, wherein the control unit
Wherein the speech recognition dictionary for voice recognition and the pronunciation dictionary for error detection are received and stored from the service device. Recognizing a word corresponding to a user's voice using a speech recognition dictionary for speech recognition that defines a reference phoneme string that can be generated for each word for speech recognition;
Detecting a pronunciation error included in the user's voice by using a pronunciation dictionary for error detection that defines an error phoneme sequence that can be generated for each word in order to detect an error in user utterance; And
If the user's voice is not matched with the standard phoneme string but the similarity is within a predetermined range and the phoneme string corresponding to the phonetic dictionary for error detection does not exist, a phoneme string corresponding to the user's voice is added to the phonetic dictionary step;
And a speech recognition unit for recognizing the speech sound. delete delete

Images