KR-102962664-B1 - AN ARTIFICIAL INTELLIGENCE APPARATUS FOR PERFORMING SPEECH RECOGNITION AND METHOD FOR THE SAME

KR102962664B1KR 102962664 B1KR102962664 B1KR 102962664B1KR-102962664-B1

Abstract

The present invention provides an artificial intelligence device comprising: a database storing correction data that replaces a predetermined voice command; a microphone receiving a first voice command from a first user; and a processor that, when an action to be performed for the first voice command is not determined, stores the first voice command in the database, obtains correction data that replaces the first voice command from a second user, maps the first voice command and the correction data, and stores the result in the database.

Inventors

원종필

Assignees

엘지전자 주식회사

Dates

Publication Date: 20260508
Application Date: 20190806

Claims (19)

In an artificial intelligence device that performs speech recognition, A database storing correction data that replaces a predetermined voice command; A microphone that receives a first voice command from a first user; A processor comprising, when an action to be performed for the first voice command is not determined, storing the first voice command in the database, obtaining a second voice command or text data as correction data replacing the first voice command from a second user, and mapping the first voice command and the correction data and storing them in the database. Artificial intelligence device.
In paragraph 1, The above processor is, Searching for a voice command having a pattern similar to the first voice command from the database, and if no voice command having a pattern similar to the first voice command is found, storing the first voice command in the database. Artificial intelligence device.
In paragraph 1, The above processor is, Acquiring the second voice command from the second user, and if the second user is determined to be a user with correction authority based on the second voice command, acquiring the second voice command as correction data that replaces the first voice command. Artificial intelligence device.
In paragraph 1, The above processor is, Acquiring the second voice command from the second user, and when an action to be performed on the second voice command is determined, acquiring the second voice command as correction data that replaces the first voice command. Artificial intelligence device.
In paragraph 1, The above processor is, Acquiring the text data from the second user, and when an action to be performed on the text data is determined, acquiring the text data as correction data that replaces the first voice command, Artificial intelligence device.
In paragraph 1, The above processor is, Obtaining correction data from the second user that modifies previously stored correction data replacing the first voice command, and obtaining the modified correction data as correction data replacing the first voice command. Artificial intelligence device.
In paragraph 1, The above microphone is, A third voice command is received from the first user, and The above processor is, Acquiring correction data that replaces the third voice command from the above database, determining an action to be performed on the correction data that replaces the third voice command, and performing voice recognition. Artificial intelligence device.
◈Claim 8 was waived upon payment of the establishment registration fee.◈ In Paragraph 7, The above processor is, Searching for a voice command having a pattern similar to the third voice command from the above database, and obtaining correction data that replaces the voice command having the searched similar pattern as correction data that replaces the third voice command. Artificial intelligence device.
◈Claim 9 was waived upon payment of the establishment registration fee.◈ In Paragraph 7, It further includes a communication unit that transmits correction data replacing the above-mentioned third voice command to an NLP server that performs intent analysis, and The above processor is, Acquiring intent analysis information from the above NLP server and performing speech recognition, Artificial intelligence device.
A speech recognition method performed by an artificial intelligence device comprising a database storing correction data that replaces a predetermined voice command, wherein A step of receiving a first voice command from a first user; If the action to be performed for the first voice command is not determined, the step of storing the first voice command in the database; A step of obtaining a second voice command or text data as correction data replacing the first voice command from a second user; and A step comprising mapping the first voice command and the acquired correction data and storing them in the database. Voice recognition method.
In Paragraph 10, The step of storing the first voice command in the database is A step of searching for a voice command having a pattern similar to the first voice command from the above database; and If a voice command having a pattern similar to the first voice command is not found, the method includes the step of storing the first voice command in the database. Voice recognition method.
◈Claim 12 was waived upon payment of the establishment registration fee.◈ In Paragraph 10, The step of obtaining correction data that replaces the first voice command is, A step of obtaining the second voice command from the second user; and If the second user is determined to be a user with correction authority based on the second voice command, the method includes the step of acquiring the second voice command as correction data that replaces the first voice command. Voice recognition method.
◈Claim 13 was waived upon payment of the establishment registration fee.◈ In Paragraph 10, The step of obtaining correction data that replaces the first voice command is, A step of obtaining the second voice command from the second user; and When an action to be performed for the second voice command is determined, the method includes the step of acquiring the second voice command as correction data that replaces the first voice command. Voice recognition method.
◈Claim 14 was waived upon payment of the establishment registration fee.◈ In Paragraph 10, The step of obtaining correction data that replaces the first voice command is, A step of obtaining the text data from the second user; and If an action to be performed on the above text data is determined, the method includes the step of acquiring the above text data as correction data that replaces the first voice command. Voice recognition method.
◈Claim 15 was waived upon payment of the establishment registration fee.◈ In Paragraph 10, The step of obtaining correction data that replaces the first voice command is, A step of obtaining correction data that modifies previously stored correction data replacing the first voice command from the second user; and The method comprising the step of obtaining the correction data to be modified above as correction data that replaces the first voice command, Voice recognition method.
◈Claim 16 was waived upon payment of the establishment registration fee.◈ In Paragraph 10, A step of receiving a third voice command from the first user; A step of obtaining correction data that replaces the third voice command from the above database; and A method further comprising the step of performing voice recognition by determining an action to be performed on correction data that replaces the third voice command. Voice recognition method.
◈Claim 17 was waived upon payment of the establishment registration fee.◈ In Paragraph 16, The step of obtaining correction data that replaces the third voice command from the above database is, A step of searching for a voice command having a pattern similar to the third voice command from the above database; and The method comprises the step of obtaining correction data that replaces a voice command having the searched similar pattern as correction data that replaces the third voice command. Voice recognition method.
◈Claim 18 was waived upon payment of the establishment registration fee.◈ In Paragraph 16, The step of performing the above-mentioned voice recognition is, A step of transmitting correction data replacing the above third voice command to an NLP server that performs intent analysis; and A method comprising the step of obtaining intent analysis information from the above NLP server and performing speech recognition, Voice recognition method.
In an artificial intelligence device that performs speech recognition, A database storing correction data that replaces a predetermined voice command; A microphone that receives voice commands from a user; and A processor comprising acquiring a second voice command or text data as correction data replacing the voice command from the above database, and determining an action to be performed on the correction data replacing the voice command to perform voice recognition, Artificial intelligence device.

Description

An artificial intelligence apparatus for performing speech recognition {AN ARTIFICIAL INTELLIGENCE APPARATUS FOR PERFORMING SPEECH RECOGNITION AND METHOD FOR THE SAME} The present invention relates to an artificial intelligence device capable of performing voice recognition by acquiring correction data that replaces voice commands. The competition in voice recognition technology, which began with smartphones, is expected to heat up in earnest inside the home, coinciding with the full-scale expansion of the Internet of Things (IoT). In particular, a noteworthy point is that the device is an artificial intelligence (AI) device capable of issuing commands and engaging in conversation via voice. Speech recognition services utilize a massive database to select the optimal answer to a user's question. The voice search function also works by converting input voice data into text on a cloud server for analysis, and then retransmitting real-time search results to the device based on the analysis. Cloud servers possess the computing capability to classify numerous words into voice data categorized by gender, age, and accent, store them, and process them in real time. However, there is a problem in processing speech recognition for voices spoken by young children who have not yet learned the language, people with strong regional dialects, or people with unclear pronunciation. In addition, there are many difficulties in generating and applying training data tailored to the characteristics of all speakers. Therefore, the need for artificial intelligence devices capable of recognizing the voices of various users is increasing. FIG. 1 shows an AI device (100) according to one embodiment of the present invention. FIG. 2 shows an AI server (200) according to one embodiment of the present invention. FIG. 3 shows an AI system (1) according to one embodiment of the present invention. FIG. 4 is a drawing for explaining a voice system according to one embodiment of the present invention. FIG. 5 is a diagram illustrating a method for collecting learning data optimized for user-specific characteristics by storing correction data for voice commands according to an embodiment of the present invention. FIG. 6 is an operation flowchart illustrating a method for storing correction data for a voice command according to an embodiment of the present invention. FIG. 7 is an operation flowchart illustrating a method for performing voice recognition using correction data for a voice command according to an embodiment of the present invention. FIGS. 8 to 10 are drawings for explaining the process of an artificial intelligence device according to an embodiment of the present invention collecting correction data for a voice command and performing voice recognition using the correction data. FIG. 11 is a diagram illustrating the process of adding or editing correction data for a voice command stored in an artificial intelligence device according to an embodiment of the present invention. Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Identical or similar components regardless of drawing symbols will be assigned the same reference number, and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably solely for the ease of drafting the specification and do not inherently possess distinct meanings or roles. Furthermore, in describing embodiments disclosed in this specification, if it is determined that a detailed description of related prior art could obscure the essence of the embodiments disclosed in this specification, such detailed description will be omitted. Additionally, the attached drawings are intended only to facilitate understanding of the embodiments disclosed in this specification; the technical concept disclosed in this specification is not limited by the attached drawings, and it should be understood that they include all modifications, equivalents, and substitutions that fall within the spirit and technical scope of the present invention. Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. Artificial Intelligence (AI) Artificial intelligence refers to the field of researching artificial intelligence or the methodologies to create