Neckband wearable device for converting speech into braille and editing same

Abstract

A neckband wearable device is disclosed. The neckband wearable device comprises: a body which can be mounted on a user's neck; a microphone which is provided in the body, recognizes speech, and converts the speech into audio signals; and a data processing unit which is embedded in the body, converts the audio signals into text by using an artificial intelligence algorithm, and converts the text into braille data.

Description

Neckband wearable device that converts and edits voice into Braille

[0001] The present invention relates to a neckband wearable device, and more specifically, to a neckband wearable device that is mounted on a user's neck and can convert the user's voice into Braille data, edit the Braille data, and output it as sound.

[0002] Braille is a tactile writing system developed by Louis Braille and designed to help the visually impaired read and write. Braille uses cells composed of six dots to represent the alphabet, numbers, and symbols, and can be perceived tactilely. As Braille can express various languages ​​and symbols, it is an important tool that enables the visually impaired to access information.

[0003] While many visually impaired individuals learn Braille at special education institutions, only a fraction of the total visually impaired population can use it fluently. Braille usage rates vary significantly depending on educational opportunities and social environments, and some studies attribute the low usage rate to the advancement of digital technology.

[0004] Meanwhile, visually impaired individuals who use Braille have opportunities for reading, writing, and accessing information, but they still face limitations in accessing information due to a lack of Braille materials.

[0005] To address these issues, it is necessary to develop and disseminate Braille-related technologies, such as Braille printers and Braille laptops. Furthermore, there is a need for technological development that supports Braille users in easily accessing diverse information by creating devices that integrate Braille and digital technologies.

[0006] The present invention provides a neckband wearable device capable of instantly converting voice data into Braille data.

[0007] In addition, the present invention provides a neckband wearable device capable of outputting Braille data as voice data.

[0008] In addition, the present invention provides a neckband wearable device capable of generating and editing Braille data.

[0009] A neckband wearable device according to an embodiment of the present invention includes: a body that can be mounted on a user's neck; a microphone provided on the body that recognizes voice and converts it into an audio signal; and a data processing unit embedded in the body that converts the audio signal into text using an artificial intelligence algorithm and converts the text into Braille data.

[0010] Additionally, it may further include a data storage unit for storing the audio signals and the Braille data; and a dial button provided at one end of the body, capable of selecting at least one of the audio signals and the Braille data stored in the data storage unit in a dial manner.

[0011] In addition, the dial button can adjust the playback position of the audio signal while the audio signal is being played.

[0012] Additionally, the microphone comprises: a first microphone provided on the upper surface of the body and generating a first audio signal by recognizing the voice; and a second microphone provided at the front end of the body and generating a second audio signal by recognizing the voice over a wider range than the first microphone, wherein the data processing unit can analyze the second audio signal to generate an ambient sound removal filter and filter the first audio signal with the ambient sound removal filter.

[0013] Additionally, the system further includes a data transmission and reception unit that transmits the braille data to an external device including a braille output device and receives braille data from the external device, wherein the data processing unit can convert the braille data received from the external device into text and audio signals.

[0014] Additionally, it may further include a speaker provided on the upper surface of the body and outputting an audio signal converted by the data processing unit toward the user's ear.

[0015] Additionally, the device further includes a biometric information recognition unit provided on the outer surface of the body and recognizing the user's biometric information, wherein the data processing unit can unlock the device by comparing the user's biometric information recognized by the biometric information recognition unit with the user's biometric information stored previously.

[0016] Additionally, the data processing unit can convert the braille data into an audio playback signal in editing mode, set an editing point designated by the user while the audio playback signal is being played, add an inserter audio signal generated through the microphone to the editing point, and convert the final audio signal, in which the inserter audio signal is inserted into the audio playback signal, into braille data.

[0017] According to the present invention, the user's spoken voice is recognized by a microphone, and the user's voice data is converted into text data and Braille data using an artificial intelligence algorithm, so the content of the user's speech can be immediately generated as Braille data.

[0018] In addition, according to the present invention, the data processing unit converts Braille data into voice data and outputs the converted voice data through a speaker, so that the user can easily check Braille information.

[0019] In addition, according to the present invention, an edit point can be selected in voice data, an insert audio signal is added to the edit point, and the data can be edited by converting it into Braille data.

[0020] In addition, according to the present invention, since the generation, editing, and output of Braille data are easy, accessibility to Braille for the visually impaired is maximized, and the problem of the digital divide can be resolved.

[0021] FIG. 1 is a perspective view showing a neckband wearable device according to an embodiment of the present invention.

[0022] FIG. 2 is a front view showing the neckband wearable device of FIG. 1.

[0023] Figure 3 is a drawing showing the wearing state of the neckband wearable device of Figure 1.

[0024] Figure 4 is a diagram showing the configuration of the neckband wearable device of Figure 1.

[0025] FIG. 5 is a diagram showing the process of generating Braille data in a data processing unit according to an embodiment of the present invention.

[0026] A neckband wearable device according to an embodiment of the present invention includes: a body that can be mounted on a user's neck; a microphone provided on the body that recognizes voice and converts it into an audio signal; and a data processing unit embedded in the body that converts the audio signal into text using an artificial intelligence algorithm and converts the text into Braille data.

[0027] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. However, the technical concept of the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided to ensure that the disclosed content is thorough and complete and to sufficiently convey the concept of the present invention to those skilled in the art.

[0028] In this specification, when a component is described as being on another component, it means that it may be formed directly on the other component or that a third component may be interposed between them. Additionally, in the drawings, the thicknesses of the films and regions are exaggerated for the effective description of the technical content.

[0029] Additionally, although terms such as first, second, third, etc., have been used to describe various components in the various embodiments of this specification, these components should not be limited by such terms. These terms are used merely to distinguish one component from another. Accordingly, what is referred to as the first component in one embodiment may be referred to as the second component in another embodiment. Each embodiment described and illustrated herein also includes its complementary embodiment. Furthermore, in this specification, "and / or" is used to mean including at least one of the components listed before and after it.

[0030] In the specification, singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, terms such as "include" or "have" are intended to specify the existence of the features, numbers, steps, components, or combinations thereof described in the specification, and should not be understood as excluding the existence or addition of one or more other features, numbers, steps, components, or combinations thereof. Additionally, in this specification, "connection" is used to include both indirectly connecting multiple components and directly connecting them.

[0031] Furthermore, in describing the present invention below, if it is determined that a detailed description of related known functions or configurations could unnecessarily obscure the essence of the invention, such detailed description will be omitted.

[0032]

[0033] FIG. 1 is a perspective view showing a neckband wearable device according to an embodiment of the present invention, FIG. 2 is a front view showing the neckband wearable device of FIG. 1, FIG. 3 is a drawing showing the wearing state of the neckband wearable device of FIG. 1, and FIG. 4 is a drawing showing the configuration of the neckband wearable device of FIG. 1.

[0034] Referring to FIGS. 1 to 4, a neckband wearable device (100) is mounted on a user's neck and converts the user's voice into Braille data. The Braille conversion device (100) receives voice data from an external device and can convert the received voice data into Braille data. The Braille conversion device (100) can transmit the converted Braille data to a Braille output device.

[0035] The neckband wearable device (100) includes a body (110), a microphone (120), a speaker (130), a biometric information recognition unit (140), a sound output button (150), a sound recording button (160), a dial button (170), a data storage unit (180), a data transmission and reception unit (190), and a data processing unit (210).

[0036] The body (110) is mounted on the user's neck. The body (110) is provided in a generally U-shape. When mounted on the user's neck, one end of the body (110) is positioned on one side in front of the user's neck, and the other end is provided on the other side in front of the user's neck. The user's neck enters the space between the one side and the other side of the body (110), and the body (110) is provided around the back of the user's neck. The body (110) is provided with a flexible material so as to be variable according to the user's neck circumference size.

[0037] A microphone (120) is provided to the body (110) and recognizes speech and converts it into an audio signal.

[0038] According to an embodiment, the microphone (120) includes a first microphone (121) and a second microphone (125).

[0039] A first microphone (121) is provided to the body (110) at a point as close as possible to the user's mouth while the user is wearing the neckband wearable device (100). A directional microphone with a relatively limited voice recognition range may be used for the first microphone (121). The first microphone (121) is provided so that the voice recognition range is directed toward the user's mouth. According to an embodiment, the first microphone (121) is provided on the upper surface of the body (110). The first microphone (120) receives the user's spoken voice relatively sensitively compared to ambient sounds. The first microphone (121) converts the received voice to generate a first audio signal.

[0040] The second microphone (125) primarily receives ambient sounds rather than the user's spoken voice. The second microphone (125) uses an omnidirectional microphone (120) that has a relatively wider voice recognition range than the first microphone (121). The second microphone (125) is provided at a location further away from the user's mouth than the first microphone (121) while the user is wearing the neckband wearable device (100). According to an embodiment, the second microphone (125) may be provided at one end of the body (110). The second microphone (125) receives the user's spoken voice and sounds generated in the surroundings while the user is speaking. Relatively speaking, sounds generated in the surroundings are received more sensitively than the user's spoken voice.

[0041] The speaker (130) outputs an audio signal to the outside. The speaker (130) is positioned close to the user's ear so that the user can easily hear the audio signal, and outputs the audio signal toward the user's ear. According to an embodiment, a super-directional speaker (130) may be used so that the audio signal can be output toward the user's ear. Alternatively, a standard speaker (130) with a wider sound output range than a super-directional speaker may be used.

[0042] A biometric information recognition unit (140) is provided on the body (110), and at least one area is exposed to the outside. The biometric information recognition unit (140) recognizes the user's biometric information. According to an embodiment, the biometric information recognition unit (140) recognizes the user's fingerprint.

[0043] The sound output button (150) is located in a part of the body (110) and is provided as a push button. When the user presses the sound output button (150), an audio signal is played and output to the outside through the speaker (130). When the user presses the sound output button (150) again, the playback of the audio signal is stopped.

[0044] The sound recording button (160) is located in a part of the body (110) and is provided as a push button. When the user presses the sound recording button (160), the voice received through the microphone (120) is stored as audio data.

[0045] The dial button (170) is located in a part of the body (110) and is provided as a dial-type button. When the user rotates the dial button (170), audio signals and Braille data stored in the data storage unit (180) are scrolled sequentially one by one, and the audio signals are automatically played in the order in which they are scrolled. Additionally, when the dial button (170) is pressed, data can be selected. The user can view and select data stored in the data storage unit (180) through the dial button (170).

[0046] The data storage unit (180) stores user information, audio data, text data, and Braille data. The user information includes the user's biometric signal stored through the biometric information recognition unit (140). The data storage unit (180) assigns an identification code to each of the audio data, text data, and Braille data, and stores the data according to the identification code. The audio data, and the text data and Braille data converted and generated therefrom, may be assigned the same identification code.

[0047] The data transmission and reception unit (190) transmits and receives various types of data. Specifically, the data transmission and reception unit (190) can transmit Braille data to a Braille output device. A Braille output device is a device that prints or outputs information in the form of Braille. The output produced by the Braille output device is printed with a character system consisting of small protrusions, which can be read by the user by touching it with their finger.

[0048] In addition, the data transmission and reception unit (190) can receive Braille data from an external device.

[0049] Additionally, the data transmission and reception unit (190) can receive audio data from a peripheral device. The received audio data is stored in the data storage unit (180). The peripheral device is a device capable of storing audio data or recording voice, and includes digital terminals capable of communicating with the data transmission and reception unit (190) via a network, such as a smartphone, laptop, tablet computer, or PDA (Personal Digital Assistant).

[0050] The data processing unit (210) is embedded in the body (110) and converts an audio signal into text using an artificial intelligence algorithm, and converts the converted text into Braille data.

[0051] According to one embodiment, the data processing unit (210) can convert audio data received from the microphone (120) into text and convert the converted text into Braille data.

[0052] According to another embodiment, the data processing unit (210) can convert audio data received from a peripheral device into text and convert the converted text into Braille data.

[0053] FIG. 5 is a diagram showing the process of generating Braille data in a data processing unit according to an embodiment of the present invention.

[0054] Referring to Fig. 5, the artificial intelligence algorithm can generate Braille data using cepstral feature extraction (211), MFCC (Mel Frequency Cepstral Coefficient, 212), Gaussian acoustic model (213), HMM Lexicon (Hidden Markov Model Lexicon, 214), text decoder (215), and Braille decoder (216).

[0055] The feature extraction method (211) extracts important features from an audio signal. The feature extraction method (211) analyzes the frequency band information of the audio signal and transforms the frequency spectrum. This helps to recognize and process speech more effectively in subsequent steps. In particular, identifying features in the frequency domain can provide basic data for subsequent processing.

[0056] The MFCC (212) models the way the human ear perceives frequencies in audio signals. The MFCC (212) converts the frequency spectrum into a Mel scale to generate a vector that reflects physiological auditory characteristics. The generated MFCC (212) effectively contains speech, which can contribute to increasing the signal recognition rate.

[0057] The Gaussian acoustic model (213) probabilistically models specific phonemes (sound units) in the audio signal. The model predicts the probability of each phoneme occurring in the audio signal and recognizes specific phonemes by reflecting the variability of the speech signal using a Gaussian distribution.

[0058] The HMM Lexicon (215) determines the probabilistic association between each phoneme and feature vector by considering the temporal change of the audio signal. The HMM Lexicon (215) determines whether a given input signal corresponds to a sequence of specific phonemes. This determination can be an important part of text conversion and Braille conversion.

[0059] A text decoder (Text Decodor, 216) converts audio data analyzed by the process described above into text data.

[0060] A Braille decoder (217) converts text data into Braille data. The Braille decoder (217) can generate Braille data by performing preprocessing of text data, character and symbol conversion, rule unification process, context recognition, and postprocessing of Braille data.

[0061] Text data preprocessing removes unnecessary spaces and special characters, converts uppercase letters to lowercase, or standardizes specific words to improve the consistency of the conversion.

[0062] Character and symbol conversion applies rules to convert each character, symbol, and number into Braille. In this process, a mapping table for Braille codes is used to convert each element of the text into its corresponding Braille.

[0063] The rule standardization process unifies rules such as line breaks and punctuation, and maintains a unified format.

[0064] Context recognition defines the meaning of words differently depending on the context and can convert Braille accordingly.

[0065] Braille data post-processing converts Braille into data by adjusting the output format or storage method of the Braille data.

[0066] Additionally, the data processing unit (210) can convert Braille data received from an external device into audio data. The converted audio data can be output externally through a speaker (130). According to one example, the Braille data received from an external device includes the content of various publications such as newspapers and books.

[0067] Additionally, the data processing unit (210) analyzes the second audio signal to generate an ambient sound removal filter and filters the first audio signal using the ambient sound removal filter. As previously described, the second audio signal sensitively receives sounds generated in the surroundings while the user is speaking. The data processing unit (210) analyzes the second audio signal to analyze the frequency band information of the surrounding sounds and extracts feature points from the corresponding frequency band information. Then, it generates an ambient sound removal filter using this information. The first audio signal mainly contains the user's spoken voice and incidentally contains surrounding sounds. Since these surrounding sounds have the same sound source as the surrounding sounds included in the second audio signal, they have the same frequency band information and feature points as the second audio signal. Accordingly, the data processing unit (210) can generate an audio signal that clearly contains the user's spoken voice by removing the voice having the same frequency band information and feature points as the ambient sound removal filter from the second audio signal. This audio signal can improve accuracy in conversion into text data and Braille data.

[0068] Additionally, the data processing unit (210) analyzes the first audio signal to generate a user voice removal filter and filters the second audio signal using the user voice removal filter. Specifically, the data processing unit (210) analyzes the first audio signal to analyze the frequency band information of the user voice and extracts feature points from the corresponding frequency band information. Then, it generates a user voice removal filter using this. Furthermore, the data processing unit (210) can generate an audio signal that clearly includes the other party's spoken voice by removing voices from the second audio signal that have the same frequency band information and feature points as the user voice removal filter. The data processing unit (210) converts the other party's audio signal into text data and Braille data.

[0069] Through the process described above, voice data of the user and the person conversing with the user can be converted into Braille data.

[0070] Additionally, the data processing unit (210) can edit audio data stored in the data storage unit (180). Hereinafter, the process of editing audio data in the data processing unit (210) is referred to as the editing mode.

[0071] When a user controls the data processing unit (210) using at least one of the aforementioned buttons, the data processing unit (210) may be switched to an edit mode. According to an embodiment, the data processing unit may be switched to an edit mode by controlling the dial button (170).

[0072] When the user selects audio data to be edited using the dial button (170), the data processing unit (210) plays the corresponding audio data. If the user presses the dial button (170) while the audio data is playing, that point is set as the editing point.

[0073] In this state, the user generates an Insert Audio Signal. The Insert Audio Signal can be generated from the user's spoken voice, or selected from a pre-stored audio signal or an audio signal received from another peripheral device.

[0074] The generated inserter audio signal is inserted at an editing point set by the user. The data processing unit (210) generates a final audio signal in which the inserter audio signal is inserted into the existing audio data. Then, the data processing unit (210) converts the final audio signal into text data and converts the text data into Braille data.

[0075] Through this process, the user can edit the audio signals stored in the data storage unit (180).

[0076]

[0077] The operation process of the neckband wearable device described above will be explained in detail below.

[0078] The user touches the biometric information recognition unit (140) with a finger while wearing the Braille conversion device around their neck. The biometric information recognition unit (140) recognizes the fingerprint information of the finger, and the data processing unit (210) compares the fingerprint information with the fingerprint information stored in the data storage unit (180) to identify whether the user is a registered user. If the fingerprint that touched the biometric information recognition unit (140) matches the fingerprint information of a previously registered user, the lock state is unlocked.

[0079] After the lock state is unlocked, the user presses the sound recording button (160). In this state, when the user converses with another person, the user's voice and the voices of people around them are received by the first microphone (120) and the second microphone (125). The user's voice is primarily recognized by the first microphone (121), and the voices of people around them are recognized by the second microphone (125). The voice recognized by the first microphone (121) is generated as a first audio signal, and the voice recognized by the second microphone (125) is generated as a second audio signal. The first audio signal and the second audio signal are stored in the data storage unit (180).

[0080] The data processing unit (210) converts audio data into text using the artificial intelligence algorithm and converts the converted text into Braille data.

[0081] First, the data processing unit (210) analyzes the second audio signal to generate an ambient sound removal filter. According to an embodiment, the data processing unit (210) analyzes the frequency band information of ambient voice included in the second audio signal and extracts feature points from the corresponding frequency band information to generate an ambient sound removal filter. Then, the first audio signal is filtered with the ambient sound removal filter to generate an audio signal that clearly includes the user's spoken voice. Since the second audio signal sensitively includes voices generated in the surroundings while the user is speaking, the frequency band information and feature points of the ambient voice can be effectively extracted. Furthermore, the first audio signal includes some of the ambient voice, and the ambient sound can be accurately removed by using the ambient sound removal filter.

[0082] Additionally, the data processing unit (210) analyzes the first audio signal to generate a user voice removal filter and filters the second audio signal using the user voice removal filter. Specifically, the data processing unit (210) analyzes the first audio signal to analyze the frequency band information of the user voice and extracts feature points from the corresponding frequency band information. Then, it generates a user voice removal filter using this. Furthermore, the data processing unit (210) can generate an audio signal that clearly includes the other party's spoken voice by removing voices from the second audio signal that have the same frequency band information and feature points as the user voice removal filter.

[0083] The data processing unit (210) generates text data from a first audio signal from which ambient sound has been removed and a second audio signal from which the user's voice has been removed, and generates Braille data from the text data.

[0084] The above-described feature extraction method (Cepstral feature extraction, 211), MFCC (Mel Frequency Cepstral Coefficient, 212), Gaussian acoustic model (Gaussian acoustic model, 213), HMM Lexicon (Hidden Markov Model Lexicon, 215), Text Decodor (Text Decodor, 216), and Braille Decoder (Braille Decoder, 217) are used to generate text data and Braille data.

[0085] When the generation of Braille data is complete, the first audio signal, the second audio signal, the text data, and the Braille data are assigned the same identification code and stored in the data storage unit (180).

[0086] When the user scrolls through audio data stored in the data storage unit (180) using the dial button (170), the scrolled audio data is played sequentially. The voice signal of the audio data is output through the speaker (130). When the user presses the sound output button (150) while listening to the voice signal, the corresponding audio data is played externally.

[0087] Additionally, when the transmission of the corresponding audio data is selected using the dial button (170), braille data having the same identification code as the corresponding audio data is transmitted to the braille output device. The braille output device outputs the braille data in braille form onto paper.

[0088] Additionally, when the user selects an edit mode using the dial button (170) and selects audio data to be edited, the corresponding audio data is played. If the user presses the dial button (170) while the audio data is playing, that point is set as the edit point. In this state, the user generates an insert audio signal. The insert audio signal can be generated from the user's spoken voice, or selected from a previously stored audio signal or an audio signal received from another peripheral device.

[0089] The generated inserter audio signal is inserted at an editing point set by the user. The data processing unit (210) generates a final audio signal in which the inserter audio signal is inserted into the existing audio data. Then, the data processing unit (210) converts the final audio signal into text data and converts the text data into Braille data.

[0090]

[0091] The aforementioned neckband wearable device instantly converts voice data into Braille data, enabling 93% of visually impaired people who cannot read Braille to easily read Braille data. In addition, it facilitates the creation and editing of Braille data, thereby maximizing accessibility to Braille for the visually impaired.

[0092] In addition, immediate usability of Braille data can be provided by playing the Braille data as sound through a speaker or by outputting it through a Braille output device.

[0093] Furthermore, it can resolve the issue of the digital divide among the visually impaired stemming from Braille accessibility. Additionally, by converting existing digital text into Braille data and audio and providing it, it can resolve the digital divide between the visually impaired and non-visually impaired.

[0094]

[0095] Although the present invention has been described in detail using preferred embodiments, the scope of the invention is not limited to specific embodiments and should be interpreted by the appended claims. Furthermore, those skilled in the art will understand that many modifications and variations are possible without departing from the scope of the invention.

[0096] The neckband wearable device according to the present invention can be mounted on a user's neck and used to convert the user's voice into Braille data, edit Braille data, and output sound.

Claims

1. A body that can be mounted on the user's neck; A microphone provided in the above body that recognizes speech and converts it into an audio signal; and A neckband wearable device embedded in the body and comprising a data processing unit that converts the audio signal into text or Braille data using an artificial intelligence algorithm.

2. In Paragraph 1, A data storage unit for storing the above audio signals and the above Braille data; and A neckband wearable device provided at one end of the body and further comprising a dial button capable of selecting at least one data among the audio signals and the Braille data stored in the data storage unit in a dial manner.

3. In Paragraph 2, The above dial button is a neckband wearable device capable of adjusting the playback position of the audio signal while the audio signal is being played.

4. In Paragraph 1, The above microphone is, A first microphone provided on the upper surface of the body and generating a first audio signal by recognizing the voice; and A second microphone provided at the tip of the body, which recognizes the voice over a wider range than the first microphone and generates a second audio signal, wherein The above data processing unit is, A neckband wearable device that analyzes the second audio signal to generate an ambient sound removal filter and filters the first audio signal with the ambient sound removal filter.

5. In Paragraph 1, The above-mentioned braille data is transmitted to an external device including a braille output device, and the above-mentioned data is received from the external device, further comprising a data transmission and reception unit. The above data processing unit is a neckband wearable device that converts Braille data received from the external device into text and audio signals.

6. In Paragraph 5, A neckband wearable device further comprising a speaker provided on the upper surface of the body and outputting an audio signal converted by the data processing unit toward the user's ear.

7. In Paragraph 1, The body further includes a biometric information recognition unit provided on the outer surface of the body and recognizing the biometric information of the user, The above data processing unit is, A neckband wearable device that unlocks by comparing the user's biometric information recognized by the above-mentioned biometric information recognition unit with the user's biometric information stored in advance.

8. In Paragraph 1, The above data processing unit is, In edit mode, the above braille data is converted into an audio playback signal, and while the audio playback signal is being played, an edit point specified by the user is set, Add the insert audio signal generated through the above microphone to the above editing point, and A neckband wearable device that converts a final audio signal, in which the inserter audio signal is inserted into the audio playback signal, into Braille data.

Images