KR20200024484A - Speech to text recording apparatus and speech-text converting system using the same - Google Patents

Abstract

The present invention relates to an STT recording device for recording a voice using a small device, converting a voice into a text, and sharing information on a voice and a text with various terminals and a voice-to-text conversion system using the same. By recording and storing a voice with optimal sound quality through a dedicated recording device with a small size and receiving and storing text information and linked information including synchronization information for a voice and a text through an external server together, utilization may be maximized by transmitting the linked information and information related to a voice file including a text file together to another terminal.

Description

Speech to text recording apparatus and speech-text converting system using the same}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speech to text (STT) recording apparatus and a speech-to-text converting system using the same. An STT recording apparatus and a voice-to-text conversion system using the same are provided.

As information society where knowledge and information are important has matured, various information exchanges and learning methods are used on a daily basis. In particular, with the rapid development of mobile devices and communication networks, information exchange methods and learning methods using multimedia contents from a remote place are becoming familiar.

However, it is still not comparable to meeting with the person to whom information is actually exchanged, such as face-to-face meetings, counseling, and lectures.

In the case of face-to-face information exchange or learning, the desired information can be obtained through faithful participation, and the memory can be revived by taking notes or notes, but the method of recording voice in a face-to-face situation in order to obtain realistic information faithfully It is actively used.

Various voice recording apparatuses exist for such voice recording, and the voice recording function exists in a smartphone or a mobile terminal owned by most people, so that voices of face-to-face situations can be recorded as needed.

Dedicated voice recording devices can efficiently record and manage voices through optimized sound quality and compression ratios, but if these contents accumulate, there are many voice files that cannot be understood quickly. Is difficult and takes a long time to understand the content. In particular, due to the limited interface, it is not easy to select a voice file containing the desired content or to find and listen to the part containing the content.

On the other hand, in the case of a voice recording function (application) of a smart phone or a portable terminal with a convenient user interface, the voice can be recorded at a necessary time in that it is always carried, but the basic function is a communication function including a voice call. It is difficult to use for voice recording only because an incoming call or a call through various applications can be received while a voice is being recorded in the system.It is not optimized for voice recording. Since insufficient storage space is occupied by voice recording, there is an inconvenience in that the recorded voice must be deleted or moved to another place.

That is, in the case of a dedicated voice recording device, a large number of voice information can be stored and played back when necessary or transferred to another terminal, but there is a limitation that it is difficult to manage or search too much voice information in association with its contents, and add a recording function. In case of using other terminal with function, recording sound quality is not good and it is difficult to accumulate and store the recording contents by the space occupied by the recording, and also it is difficult to check and manage the contents or search for desired contents. have.

Therefore, it is possible to carry and accumulate a large amount of recorded voices with the optimal sound quality, and to easily search for the contents and play the desired voice file. There is a need for a new dedicated recording device and a voice-to-text conversion system using the same.

Meanwhile, according to the development of speech to text (STT) technology for converting speech into text, various functions of converting speech into text are utilized, and converting recorded speech into text can be easily performed. However, even when a human hears, the contents of the voice cannot be accurately distinguished by the pronunciation or the surrounding environment, and the pronunciation is similar, but since there are other voices with actual characters, the result of the STT is not completely reliable. . Thus, a problem may occur in that the reliability of data may be lowered by simply converting a voice into text but providing unreliable information.

After all, it is not only to record and manage voice with optimal sound quality, but to convert it to text to provide additional information, and to provide reliable information in various ways. The situation is not appearing.

Korean Patent Laid-Open Publication No. 10-2017-0046958, [A method for executing a function using an electronic device and speech recognition thereof] Korean Patent Laid-Open No. 10-2014-0006503, [Method and Apparatus for Recording and Reproducing User Voice of Mobile Terminal]

An object of the embodiments of the present invention for improving the above problems is to provide a compact dedicated recording device that can be easily and easily directed to the sound source, recording and storing the voice in the optimum sound quality and to a server configured in the communication network Receives the result according to the STT after transmission and matches with the recorded voice, and also receives and saves the interlocking information including the synchronization information for the voice and the text, so that the voice file containing the desired contents can be easily searched based on the text. The present invention provides an STT recording apparatus and a voice-to-text conversion system using the same, which allow a user to check and play a corresponding position, or to transmit information related to a voice file including such interworking information and a text file to another terminal.

In addition, another object of the embodiments of the present invention is to generate a result according to the STT by the server receiving the recorded voice from the recording device, along with the interworking information including the synchronization information for the voice and text and the reliability information for the text conversion portion STT recording device that makes it possible to easily play the voice of the part with low reliability through interworking information when the recording device or other terminal receiving related information uses a text file corresponding to the voice file by generating and transmitting it to the recording device. It is to provide a speech-to-text conversion system.

In addition, another object of the embodiments of the present invention is a voice region of a low reliability part based on the linkage information corresponding to the voice file by selecting a voice file to be deleted if the storage capacity of the recording device is insufficient or the user requests it. You can reduce the storage capacity of the voice file by creating a new voice file selected and deleting the original voice file, and play the text file corresponding to the voice file through the Text to speech (TTS) function. The low reliability part is to provide an STT recording device and a voice-to-text conversion system using the same, so that the capacity and reliability can be satisfied by using a voice file stored separately.

In order to achieve the above object, the STT recording apparatus according to an embodiment of the present invention collects a voice signal through a microphone and converts it into a digital signal, a wireless communication unit for communicating with an external server, and through a connector A wired communication unit that receives charging power and exchanges information with an external terminal, a user interface unit that receives a user input and informs the user of the state, and a storage unit that stores voice information based on the digital signal converted by the voice input unit. And providing the voice information stored in the storage to the external server through the wireless communication unit, and receiving interworking information including text information corresponding to the voice information and synchronization information of the voice information and the text information from the external server. Voice-text management unit for storing in the user, and receives the user input through the user interface unit Control the wireless communication unit and the storage unit according to the request of the voice-text management unit, and provide the storage information of the storage unit to the external terminal through the wired communication unit or store information received from the external terminal in the storage unit. And a control unit for storing.

As an example, the interworking information may include synchronization information about a reproduction time reference position or a data reference position and reliability information of the converted text with respect to the text converted corresponding to the voice information.

As one example, the interworking information may be provided separately from the text information or mixed with the text information in a tag form.

For example, the controller selects voice information to be deleted when the user inputs through the user interface unit or when the storage unit runs out of storage space according to a predetermined criterion, and corresponds to a text area having a reliability lower than a reference value in interworking information corresponding to the voice information. The voice information area may be separated to generate additional supplementary voice information, and then the corresponding voice information may be deleted.

For example, the controller selects the voice information to be deleted when the user inputs through the user interface or when the storage unit runs out of storage space according to a predetermined criterion, and provides the voice information to be deleted to the external server through the wireless communication unit. In the interworking information corresponding to the information, the voice information area corresponding to the text area having a reliability lower than the reference level may be separated to receive supplementary voice information generated separately, and the voice information may be replaced with the supplemental voice information.

For example, when the voice information area corresponding to the text area having a reliability lower than the reference is separated, the voice information area may be extended and separated up to a preset number of words or sentences existing before and after the reference based on the corresponding voice information area. It may be.

For example, when the controller selects and provides the voice information from the storage unit according to the input of the user interface unit, the controller may further include a playback unit for playing the voice information, and the playback unit may reproduce the voice information or use text corresponding to the voice information. Speech information can be provided using a text to speech (TTS) method, and when the speech information is provided using the TTS method, if a low-conversion portion is detected through interworking information corresponding to the speech information, the corresponding speech information area is replaced with text. Can play.

For example, the controller may search and select the voice information stored in the storage unit based on the content of the text when the search request including the search word received through the user interface unit, the wireless communication unit, or the wired communication unit.

The voice-to-text conversion system using the STT recording apparatus according to another embodiment of the present invention stores the collected voice information, and transmits the corresponding text information, synchronization information, and reliability information received after transmitting the voice information to the service server. STT recording device that matches the included interlocking information to the voice information and stores it in the storage unit and provides a voice signal, corresponding text information and interlocking information to the connected external terminal, and converts the voice information received from the STT recording device into text And a service server for generating interlocking information including synchronization information on a reproduction time reference position or a data reference position and reliability information of the converted text, and transmitting the text information and the interlocking information to the STT recording device.

As an example, the STT recording apparatus may separately generate a voice information area corresponding to a text area having a reliability lower than a reference from interlocking information by selecting the voice information to be deleted when the storage space of the storage unit according to a user's input or a predetermined standard is insufficient. The supplementary voice information may be alternatively stored, but the supplementary voice information may be generated by the STT recording apparatus or the service server may generate supplemental voice information about the voice information selected by the STT recording apparatus and provide the supplementary voice information to the STT recording apparatus.

The STT recording apparatus and the voice-to-text conversion system using the same according to embodiments of the present invention record and store the voice at an optimum sound quality through a small dedicated recording apparatus, and synchronize text information and voice and text through an external server. By receiving and storing the interworking information including the information, there is an effect that can maximize the utilization by transmitting the interworking information and voice file-related information including the text file to the other terminal.

According to embodiments of the present invention, an STT recording apparatus and a voice-to-text conversion system using the STT recording apparatus also generate interworking information including synchronization information on voice and text and reliability information on a text conversion portion, and then, to the recording apparatus. By transmitting, when the recording device or another terminal provided with the related information uses a text file corresponding to the voice file, it is possible to increase the reliability by facilitating the voice reproduction for the low reliability part through the interworking information.

According to embodiments of the present invention, an STT recording apparatus and a voice-to-text conversion system using the same may provide interworking information corresponding to a corresponding voice file by selecting a voice file to be deleted if the storage capacity of the recording device is insufficient or when a user controls it. By creating a new voice file that selects only the voice region of the low-reliability part as a reference and deleting the original voice file, the storage capacity for the voice file can be reduced, and the voice is also provided through the text to speech (TTS) function. The part having low reliability while playing the text file corresponding to the file has the effect of satisfying both capacity and reliability by using a separately stored voice file.

1 is a block diagram showing a voice-to-text conversion system using an STT recording apparatus according to an embodiment of the present invention.
2 shows an example of an STT recording apparatus according to an embodiment of the present invention.
3 is a block diagram of an STT recording apparatus according to an embodiment of the present invention.
4 is a configuration diagram of a service server according to an embodiment of the present invention.
5 is a conceptual diagram illustrating a characteristic operation method of the speech-to-text conversion system according to an embodiment of the present invention.
6 is a flowchart illustrating an operation process of an STT recording apparatus according to an embodiment of the present invention.
7 is a flowchart illustrating an operation of a service server according to an embodiment of the present invention.
8 is a conceptual diagram illustrating a capacity reduction method according to an embodiment of the present invention.
9 is a flowchart illustrating an operation of a capacity reduction method according to an embodiment of the present invention.

The present invention as described above will be described in detail with reference to the accompanying drawings and embodiments.

Technical terms used in the present invention are merely used to describe particular embodiments, it should be noted that it is not intended to limit the present invention. In addition, the technical terms used in the present invention should be interpreted as meanings generally understood by those skilled in the art unless the present invention is defined in any other meaning in the present invention, and is excessively comprehensive. It shall not be construed in the sense of or in the sense of being excessively reduced. In addition, when the technical terminology used in the present invention is an incorrect technical term that does not accurately express the spirit of the present invention, it should be understood as being replaced by a technical term that can be properly 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 the context before and after, and should not be interpreted in an excessively reduced sense.

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

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

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, and the same or similar components will be given the same reference numerals regardless of the reference numerals and redundant description thereof will be omitted.

In addition, in describing the present invention, when it is determined that the detailed description of the related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, it should be noted that the accompanying drawings are only for easily understanding the spirit of the present invention and should not be construed as limiting the spirit of the present invention by the accompanying drawings.

In particular, the other terminals described in the present invention are a smart phone having a wired communication function, a portable terminal, a mobile terminal, a personal digital assistant (PDA), a PMP ( It includes various terminals such as a portable multimedia player terminal, a personal computer, a notebook computer, a tablet PC, and the like, and the service server performs a function set through a web server and an application that provide a web page interface. It includes a physical or virtual server including a web application server.

Hereinafter, embodiments of the present invention will be described with reference to FIGS. 1 to 9.

1 is a block diagram showing a voice-to-text conversion system using an STT recording apparatus according to an embodiment of the present invention. As shown in the figure is small and portable STT recording device 100 for storing the voice of the speaker (1) in the optimal sound quality and matching the corresponding text information received after transmitting the corresponding voice information to the service server 200 and The service server 200 is connected to the communication network 20 and performs a text change process on the voice information recorded in association with the STT recording apparatus 100 and then provides the STT recording apparatus 100 with a result including the converted text. ).

The STT recording apparatus 100 may transmit voice information and text information corresponding thereto stored in a wired communication with other terminals 31 and 32, and the other terminals 31 and 32 may receive the received voice information and text information. It may be used together or separately, and may receive related information after accessing the service server 200 through member authentication.

That is, the service server 200 may store the received voice information and the corresponding converted text information corresponding to the member and provide the corresponding information to the other terminals 31 and 32 connected through member authentication. To this end, the service server 200 may have a member management unit 230 therein to perform member registration and authentication of a member, and service related information including voice information and converted text information received for the member. The information can be provided to the retained and certified members.

Meanwhile, in the exemplary embodiment of the present invention, the STT recording apparatus 100 and the service server 200 do not simply deliver the voice information and provide only the result of converting the voice information into text through a known voice recognition technology. Additional information can be created and used.

First, the service server 200 performs text conversion on the voice information through the internal STT conversion unit 240, and synchronizes information on the playback time reference position or data reference position with respect to the text converted in correspondence with the voice information. After generating the can be provided to the STT recording device 100 as interworking information.

Since the STT recording apparatus 100 can easily grasp the voice information and the location including the keyword input based on the text information and the interworking information, the STT recording device 100 can selectively provide the voice information including the desired content among the numerous voice information. , Can be played directly from the location containing the relevant keyword.

In addition, such interworking information may be provided separately from the text information or mixed in the form of a tag in the text information. When the audio information is played like a subtitle, the corresponding text may be displayed, and the voice played like the lyrics of a song. You can also have text displayed in response to information. This is convenient because other terminals 31 and 32 provided with the corresponding information can be executed through a general multimedia playback application.

As another additional information, the service server 200 may perform the text conversion on the voice information through the STT conversion unit 240, generate the reliability information for the text conversion portion and provide it as interworking information. May be set corresponding to the recognition rate among the results of the conversion obtained during the conversion, and the configuration of the information may vary.

For example, it is possible to generate information that numerically expresses the reliability of each recognized word, such as synchronization information, for the converted text. , Italic, underline, size, etc.), and reliability as tag information.

Since the reliability information is generated together with the synchronization information, the parts having low reliability can be easily selected to quickly select and reproduce the voice information of the corresponding parts. In this way, when you want to use the text output, the user can correct the wrong part by quickly checking the low reliability part.

On the other hand, such reliability information and corresponding speech information can be utilized as an important part to increase the reliability of the converted text, and also have long playback time and capacity when there is insufficient storage space or when files need to be cleaned up. It is possible to maintain reliability while reducing the capacity by replacing the original voice information by generating separate supplementary voice information by separating only the voice information part which is not reliable and cannot confirm the exact conversion by text alone without storing all this large voice information. have.

In this case, the timing of interworking voice information in the reliability or synchronization information may be changed to that of supplementary voice information. Alternatively, the reproduction lengths of the supplementary voice information and the original voice information may be the same, but not selected (high reliability). The subfields generate supplemental voice information with the compression ratio increased by deleting the voice information, so that the playback time is the same but the storage capacity can be significantly reduced.

Hereinafter, the operation of the present invention will be described through a more detailed configuration of the embodiments of the present invention.

Figure 2 shows an example of the STT recording apparatus according to an embodiment of the present invention, Figure 3 shows a block diagram of the STT recording apparatus according to an embodiment of the present invention.

As shown in FIG. 2, the STT recording apparatus 100 has a small size that can be easily carried and includes a directional microphone 121 and an input interface 131 for receiving a user input. A storage interface 181 for storing a removable storage unit, a connector 141 for connecting with a wired communication unit, a connector 191 for connecting a speaker or earphone / headphone for playback, and an output interface for displaying a state 132 ) Is configured.

Although not shown, the output interface 132 may further include a display unit capable of displaying text or graphics, and may be directly connected to a wired communication connector of another terminal as a memory stick, in addition to the wired communication connector 141 shown. It may be provided with a terminal (for example, various types of USB terminal).

3 is a block diagram showing the internal configuration of the STT recording apparatus 100, the voice input unit 120 to collect the voice signal through a microphone and convert it into a digital signal as shown, and performs the STT function as an external server The wireless communication unit 110 to communicate with the service server, the wired communication unit 140 receives the charging power through the connector and exchanges information with the external terminal, and a user interface unit for receiving the user's input and informs the user of the status ( 130, the storage unit 180 storing voice information based on the digital signal converted by the voice input unit 120, and the voice information stored in the storage unit 180 through the wireless communication unit 110 to the service server. And receiving interworking information including text information corresponding to the voice information from the service server and synchronization information of the voice information and the text information, and storing the received information in the storage unit 180. Charge the battery through the voice-text management unit 150, the playback unit 190 for reproducing the voice information stored in the storage unit 180, and the charging power provided through the wired communication unit 140 or a separate external power source. And converts the voice information converted into a digital signal through the power management unit 170 and the voice input unit 120 into a file format in a predetermined format and stores the stored power in the storage unit 180, and the user interface unit ( Receives a user's input through the 130 and displays the status thereof, and controls the wireless communication unit 110 and the storage unit 180 according to the request of the voice-text management unit 150, wired communication unit 140 The controller 160 may provide the storage information of the storage unit 180 to the external terminal through the storage unit 180 or store the information received from the external terminal in the storage unit 180.

Of course, the controller 160 additionally checks the power state through the power manager 170 and informs the user of the power state through the user interface 130 or the storage unit 180 according to the input of the user interface 130. The stored voice information or text information may be selected and played through the playback unit 190, and the screening process or the playback process may be displayed through the user interface 130.

The voice input unit 120 may include an internally built-in microphone, an amplifier, a filter, an analog-digital converter, and the like. The voice input unit 120 may include a wireless microphone interworking using short-range wireless communication that may be configured in the wireless communication unit 110. It can also be used selectively.

The user interface 130 includes at least one of a plurality of input means including a plurality of input switches, a touch switch, a jog shuttle, a slide switch, a rotary switch, and a volume, a light-emitting diode (LED), and a liquid crystal display. (Liquid Crystal Display: LCD), Thin Film Transistor-Liquid Crystal Display (TFT LCD), Organic Light-Emitting Diode (OLED), Flexible Display, 3D Display (3D) At least one of various output means including a display or a speaker such as a display, an e-ink display, and a light emitting diode (LED) may be implemented.

The wireless communication unit 110 may basically support various broadband wireless communication that can exchange information with a service server configured in a communication network or a short range wireless communication method for connecting to the Internet, and support a short range wireless communication method for connecting with a wireless microphone. Can be. For example, Wireless LAN (WLAN), Wireless Broadband (Wibro), WiMAX (World Interoperability for Microwave Access (Wimax), High Speed Downlink Packet Access (HSDPA), IEEE 802.16, Long Term Evolution Evolution (LTE), Wireless Mobile Broadband Service (WMBS), and Long Range (LoRa), which is a long-range IoT communication method, may be included. Meanwhile, short-range communication technologies include Bluetooth, Wi-Fi, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), UWB (Ultra Wideband), ZigBee, and Neighboring. Near Field Communication (NFC), Ultrasonic Communication (Ultra Sound Communication (USC)), Visible Light Communication (VLC), Bluetooth Low Energy (BLE), and the like may be included.

The wired communication unit 140 may use various communication methods such as power line communication (PLC), USB communication, Ethernet, IEEE 1394, and serial communication, but are highly compatible and capable of delivering power. It is preferable to use a USB communication system.

For example, when the Wi-Fi connection is possible, the voice-text manager 150 may provide voice information stored in the storage 180 to the service server through the wireless communication unit 110 and correspond to the voice information from the service server. The interlocking information including the text information and the synchronization information of the corresponding voice information and the text information is received and stored in the storage unit 180. The interlocking information may further include reliability information in addition to the synchronization information.

The controller selects voice information to be deleted when the user inputs through the user interface unit 130 or when the storage unit 180 runs out of storage space according to a predetermined criterion so that the voice information corresponding to the text area whose reliability is lower than the reference in the interlocking information is selected. The area can be separated and replaced with the supplementary voice information generated separately. That is, instead of deleting the voice information having a large storage capacity, it is possible to save the storage space by storing supplemental voice information stored by separating only the voice information of the portion having low text conversion reliability.

Such supplementary voice information may be generated by the controller 160 directly or provided through a service server.

For example, the controller 160 selects the voice information to be deleted, separates the voice information area corresponding to the text area whose reliability is lower than the reference from the interworking information corresponding to the voice information, and generates additional supplementary voice information. The voice information may be deleted and the generated supplemental voice information may be stored in the storage unit 180.

As another example, the controller 160 provides the voice information to be deleted to the service server through the wireless communication unit, and separately generates a voice information area corresponding to a text area having a reliability lower than a reference from interworking information corresponding to the voice information. After the supplementary voice information is received, the voice information may be replaced with the supplementary voice information.

In each of these cases, the supplementary speech information is changed to supplementary speech information instead of the original speech information, so that the corresponding linkage information with the synchronization information is changed according to the supplementary speech information, or the supplementary speech information has the same reproduction length as the original speech information, but the reliability is high. The interworking information can be kept intact while reducing the size of the voice information by making the actual data exist only in the low voice information area and no other data.

In addition, in the case where the voice information area corresponding to the text area whose reliability is lower than the reference is separated, if the voice information area having low reliability is separated, it is highly likely that the voice information area may not be accurately distinguished even by the human hearing. Therefore, the speech information region can be extended and separated from a predetermined number of words or sentences that exist before and after the corresponding speech information region to identify context or before and after to identify or estimate a speech having a low reliability. Make sure

On the other hand, the playback unit 190 reproduces the voice information stored in the storage unit 180 under the control of the control unit 160, or by using the text corresponding to the speech information in the text to speech (TTS) method When the voice information is provided by the TTS method, when the transform part having low reliability is detected through the interworking information corresponding to the voice information, the corresponding voice information area may be played instead of the text. That is, through the TTS, the user can listen to the voice corresponding to the stored voice information at a desired speed and tone, and the low-reliability text portion can reproduce the corresponding voice information portion instead of the text to prevent the loss of reliability due to an incorrect conversion. For example, this may be performed when all original voice information is provided or when the original voice information is deleted and only supplementary voice information is provided to reduce the capacity. In other words, even if the original voice information is deleted, reliable voice can be reproduced.

The controller 160 displays the contents of the text corresponding to the voice information stored in the storage unit 180 when the search request including the search word received through the user interface 130, the wireless communication unit 110, or the wired communication unit 140. Searching and selecting based on the criteria, it can be provided quickly and easily the voice information including the content desired by the user.

As a result, a large amount of voice information can be stored in a limited storage through the above configuration. For example, a large amount of voice information or text information can be stored in a limited storage such as storing all of a semester's lecture content or a year's meeting contents. It can be stored and can be quickly searched and checked for desired contents, and only necessary information can be selected and provided to a desired terminal.

4 is a configuration diagram of a service server according to an embodiment of the present invention.

As shown, the service server 200 is a communication unit 210 for communicating with the STT recording apparatus 100 through a communication network, and a voice information receiving unit 220 for receiving voice information transmitted from the STT recording apparatus 100. And perform member registration and authentication of the corresponding member (traditional ID / PW method (web access) or device identification information of the STT recording device 100), and permit use of the STT conversion function, Member management unit 230 and the voice information receiving unit 220 to enable the use of the text generation unit for recognizing and converting the voice to the text and the synchronization information based on the location of the playback time or capacity while generating the text And an STT converter 240 having an interlocking information generating unit for generating interlocking information including the conversion reliability information, and the text and the interlocking information converted by the STT converting unit 240 as result information. A member for storing the STT result generator 250 and the voice information received from the voice information receiver 220 and the text and interworking information generated by the STT result generator 250 for each member. The voice information storage unit 260 is provided.

In addition, the service server 200 receives the deletion target voice information itself or a file name or identification information of the corresponding deletion target voice information from the STT recording device, and provides the corresponding interworking information corresponding to the member voice information storage unit 260. In the interworking information corresponding to the corresponding voice information, the voice information area corresponding to the text area having the reliability lower than the reference level may be separated to generate supplementary voice information generated separately, and then provided to the STT recording apparatus.

5 is a conceptual diagram illustrating a characteristic operation method of the voice-to-text conversion system according to an embodiment of the present invention. As shown, the STT recording apparatus 100 records voice information recorded through the voice-text manager 150. To the service server 200, the service server 200 converts the corresponding voice information into text through the STT conversion unit 240 to generate interworking information including synchronization information and reliability information, STT result generator The corresponding information is provided to the voice-text management unit 150 of the STT recording apparatus 100 through the 250, and the controller 160 stores the received text information and the interworking information in the storage unit 180.

The service server 200 may store the received voice information, the generated text, and the interlocking information in the voice information storage unit 260 for each member, and if necessary, provide the same to the STT recording apparatus 100 again or to the registered user. It may be provided to another terminal.

Meanwhile, in some cases, the service server 200 may generate and provide supplemental voice information obtained by separating only the voice information area having low reliability after receiving the voice information. Depending on the setting, the STT recording apparatus 100 may provide original voice information. Instead, only supplementary voice information may be stored along with text and interworking information.

6 is a flowchart illustrating an operation of an STT recording apparatus according to an exemplary embodiment of the present invention. As shown, the STT recording apparatus receives a voice signal through a built-in microphone or a connected wireless microphone and digitally converts it to a storage unit. Is stored as voice information, and the voice information is transmitted to a web server (service server) existing on the web.

After receiving corresponding text information and interworking information from the corresponding web server (service server) and storing it in association with the pre-stored voice, you can use the stored voice information, text information through the built-in playback unit and the desired other terminal connected Information can be selected and delivered.

The other terminal may use voice information, text information, and interworking information, respectively, use them as multimedia information including subtitles, or use sound source information with lyrics.

7 is a flowchart illustrating an operation of a service server according to an exemplary embodiment of the present invention. As illustrated, the service server receives voice information from an STT recording apparatus and performs STT based on the corresponding voice information to perform text information. Synchronization information is generated while generating synchronization information and reliability information.

The generated text information and the interworking information may be provided to the STT recording apparatus again, and the received voice and the generated text and the interworking information may be matched and stored and then provided to the authenticated member. Of course, it can be provided to other members designated by an authenticated member, or can transmit voice, text, and interlocking data stored to a desired target (e-mail, phone number, P2P, FTP, etc.).

FIG. 8 is a conceptual view illustrating a capacity reduction method according to an embodiment of the present invention. As shown in FIG. 8, voice information, text information corresponding thereto, and interworking information are matched and stored in a storage unit of an STT recording apparatus. Here, the text information and the interworking information may exist separately or the interworking information may be integrated as one file.

When recording lectures or meetings as shown in FIG. 8A, voice information of a very long time (20 minutes 40 seconds to 59 minutes 21 seconds in the illustrated example) is stored, which takes up a considerable amount of capacity even when compressed and optimized for voice. Done.

Therefore, when the capacity of the storage unit is insufficient or the user requests compression for a predetermined capacity or more, the selected part (three voice information having the longest time in the illustrated example) is selected, and the control unit or service server of the STT recording device By checking the reliability information included in the interworking information corresponding to the selected voice information, the voice information area corresponding to the text area having the reliability below the standard is separated to generate supplemental voice information, and the STT recording apparatus corresponds to the supplementary voice information. It is stored in the storage instead of the voice information.

As shown in FIG. 8B, supplemental speech information generated by extracting speech regions having low text conversion reliability (or regions expanded to further include a front and rear part based on the speech region) from the existing speech information has a size that is the same. It will be greatly reduced.

FIG. 9 is a flowchart illustrating an operation of a capacity reduction method according to an embodiment of the present invention. As shown, when the STT recording apparatus selects a voice file to be deleted, the reliability of the text conversion is low by analyzing the interworking information on the corresponding voice file. After the voice region is selected, supplemental voice information is generated.

Such supplementary voice information may be generated by the STT recording apparatus or requested to be generated through the service server, and the result may be received by the STT recording apparatus.

The STT recording apparatus stores the supplementary voice information instead of the corresponding voice information after deleting it, and corrects the interworking information according to the supplementary voice information if necessary.

Through this process, the STT recording device can store a large amount of voice information, text information, and interworking information with limited storage capacity, and can easily and quickly select a desired object, play back as needed, or deliver desired information to other terminals. .

In the above described and illustrated with respect to preferred embodiments according to the present invention. However, the present invention is not limited to the above-described embodiments, and various modifications can be made by those skilled in the art without departing from the scope of the present invention, which is appended in the claims. .

100: STT recording device 110: wireless communication unit
120: voice input unit 121: microphone
130: user interface unit 131: input interface
132: output interface 140: wired communication unit
141: connector for wired communication 150: voice-text management unit
160: control unit 170: power management unit
180: storage unit 181: storage interface
190: playback unit 191: earphone / headphone connector
200: service server 210: communication unit
220: voice information receiving unit 230: member management unit
240: STT conversion unit 250: STT result generation unit
260: voice information storage unit for each member

Claims (10)

A voice input unit for collecting a voice signal through a microphone and converting the voice signal into a digital signal;
A wireless communication unit communicating with an external server;
A wired communication unit receiving charging power through a connector and exchanging information with an external terminal;
A user interface unit for receiving a user input and informing a user of a status;
A storage unit for storing voice information based on the digital signal converted by the voice input unit;
Providing the voice information stored in the storage unit to the external server through the wireless communication unit, and receives the text information corresponding to the voice information and the interlocking information including the synchronization information of the voice information and the text information from the external server A voice-text manager for storing in the storage;
Receives the user's input through the user interface unit and controls to display the status thereof, controls the wireless communication unit and the storage unit according to the request of the voice-text management unit, and provides the storage information of the storage unit to the external terminal or via the wired communication unit STT recording device including a control unit for storing the information received from the terminal to the storage unit.
The STT recording apparatus according to claim 1, wherein the linkage information includes synchronization information about a reproduction time reference position or a data reference position and reliability information of the converted text with respect to the converted text corresponding to the voice information.
The STT recording apparatus according to claim 1, wherein the interworking information is provided separately from the text information or mixed with the text information in a tag form.
The text area of claim 1, wherein the control unit selects voice information to be deleted when the user inputs through the user interface unit or when the storage unit runs out of storage space according to a predetermined criterion, and the reliability is less than or equal to the reference in the linkage information corresponding to the voice information. STT recording device, characterized in that to separate the voice information area corresponding to the to generate additional supplementary voice information and then delete the corresponding voice information.
The method of claim 1, wherein the controller selects the voice information to be deleted when the user inputs through the user interface or the storage unit lacks a storage space according to a predetermined criterion, and provides the voice information to be deleted to the external server through the wireless communication unit. STT, characterized in that the voice information area corresponding to the text area of which reliability is lower than a reference from the interworking information corresponding to the voice information is received, and separately generated supplementary voice information is received and the corresponding voice information is replaced with the supplemental voice information. Recording device.
The voice information area of claim 4 or 5, wherein when the voice information area corresponding to the text area whose reliability is lower than the reference is separated, up to a predetermined number of words or sentences existing before and after the voice information area based on the corresponding voice information area. STT recording device, characterized in that for expanding.
The method of claim 1, further comprising a playback unit for reproducing the voice information when the control unit selects and provides the voice information in accordance with the input of the user interface unit,
The reproducing unit may reproduce the voice information or provide the voice information in a text to speech (TTS) manner by using text corresponding to the voice information, and when the voice information is provided in the TTS manner, STT recording apparatus, characterized in that for reproducing a corresponding voice information region instead of text when the conversion portion having a low reliability is detected through the interworking information.
The STT of claim 1, wherein the control unit searches and selects voice information stored in the storage unit based on the content of the corresponding text when a search request including a search word received through a user interface unit, a wireless communication unit, or a wired communication unit is selected. Recording device.
An external terminal for storing the collected voice information, matching the received text information after receiving the voice information to the service server, and interlocking information including synchronization information and reliability information, matching the voice information, and storing the voice information in a storage unit. An STT recording apparatus for providing a voice signal, corresponding text information, and interworking information to the apparatus;
STT recording the text information and the interworking information by generating interworking information including the synchronization information of the reference time position or the data reference position and the reliability information of the converted text while converting the voice information received from the STT recording device into text. Voice-to-text conversion system using an STT recording device comprising a service server for transmitting to the device.
10. The apparatus of claim 9, wherein the STT recording apparatus selects voice information to be deleted when the storage unit runs out of storage space according to a user's input or a predetermined standard, and separates a voice information area corresponding to a text area having a reliability lower than a reference from the linkage information. Substitute and save with supplementary voice information created separately,
The supplementary speech information is generated by the STT recording apparatus or the service server generates supplementary speech information on the speech information selected by the STT recording apparatus and provides the supplementary speech information to the STT recording apparatus. system.

Images