TECHNICAL FIELD
[0001]
The present disclosure relates to an artificial intelligence server that may receive personalized information from a user requested recognition and perform personalized recognition using the received personalized information to save a storage space of the server and provide a personalized recognition service.
BACKGROUND ART
[0002]
Artificial intelligence is a field of computer engineering and information technology that studies a method for allowing computers to think, learn, self-develop, and the like that may be performed by human intelligence. The artificial intelligence means that the computers may imitate the human intelligence.
[0003]
Further, the artificial intelligence does not exist by itself, but directly or indirectly related to other fields of the computer science. Particularly in the modern age, attempts to introduce artificial intelligence elements in various fields of the information technology and to utilize the artificial intelligence elements in solving problems in the field are being actively carried out.
[0004]
In one example, technologies for recognizing and learning a surrounding situation using the artificial intelligence and providing information desired by a user in a desired form or performing an operation or function desired by the user have been actively studied.
[0005]
An electronic device providing such various operations and functions may be referred to as an artificial intelligence device.
[0006]
The artificial intelligence device may include an electronic device used directly by the user and an artificial intelligence server that provides a recognition service to the electronic device while communicating with the electronic device.
[0007]
Further, the artificial intelligence system may be composed of a plurality of electronic devices and the artificial intelligence server.
[0008]
In an artificial intelligence system, the electronic device receives input data (video, speech, or the like) and transmits the input data to the artificial intelligence server. Then, the artificial intelligence server uses the input data to acquire a recognition result (e.g., presence or absence of an obstacle, a structure thereof, what is an object in the video, or the like in case of the video, and text corresponding to the speech, meaning of the speech, or the like in case of the speech) and then transmits the recognition result to the electronic device.
[0009]
The artificial intelligence server provides the recognition service to the plurality of electronic devices. Further, generally, the artificial intelligence server obtains the recognition result by inputting the input data received from the plurality of electronic devices into one deep learning model.
[0010]
In one example, a usage environment of each of the plurality of electronic devices may be different from each other. For example, in case of speech recognition, a first electronic device may be used by a user using a dialect of a specific region, a second electronic device may be used by a user uniquely pronouncing some words, a third electronic device may be used in an environment with a lot of ambient noise, and a fourth electronic device may be used by a user who utters at a high speed.
[0011]
Further, when the recognition service is provided using one deep learning model for the plurality of electronic devices having different usage environments, it is difficult to provide a recognition service that is optimized for the usage environment of each electronic device, that is, a personalized recognition service.
[0012]
In one example, there is a scheme of operating a deep learning model corresponding to each electronic device for optimizing the recognition service to the usage environment of each electronic device. For example, the scheme is in a manner of separately operating a first deep learning model suitable for the first electronic device of the user using the dialect of the specific region, a second deep learning model suitable for the second electronic device of the user uniquely pronouncing some words, a third deep learning model suitable for the third electronic device used in the environment with a lot of ambient noise, and a fourth deep learning model suitable for the fourth electronic device of the user who utters at high speed.
[0013]
However, this scheme has a disadvantage in that a storage space required for the server becomes very large. For example, when a size of a deep learning model for the speech recognition is 80 MB and 130,000 electronic devices are provided with the speech recognition service from the server, 1 TB of the storage space will be required. In addition, considering the number of models of a product and a special case for each individual (fast uttering speed, noise environment, or the like), the storage space actually required increases significantly.
DISCLOSURE Technical Purpose
[0014]
The present disclosure is to solve the above-mentioned problems. A purpose of the present disclosure is to provide an artificial intelligence server that may receive personalized information from a user requested recognition and perform personalized recognition using the received personalized information to save a storage space of the server and provide a personalized recognition service.
Technical Solution
[0015]
In one aspect, an artificial intelligence server includes a communicator in communication with at least one electronic device and a processor for receiving input data from a specific electronic device, applying personalized information corresponding to the specific electronic device to a recognition model, inputting the input data into the recognition model having the personalized information applied thereto, thereby to obtain a final result value, and transmitting the final result value to the specific electronic device.
[0016]
In another aspect, an artificial intelligence electronic device includes a communicator in communication with an artificial intelligence server and a processor for transmitting input data to the artificial intelligence server, receiving an intermediate result value, an output value of a common model for the input data, from the artificial intelligence server, and inputting the intermediate result value to a personalized model to obtain a final result value.
Technical Effect
[0017]
According to the present disclosure, the storage space of the artificial intelligence server may be saved while providing the personalized recognition service to each of the plurality of electronic devices.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018]
FIG. 1 a illustrates an AI apparatus 100 according to an embodiment of the present disclosure.
[0019]
FIG. 1 b illustrates an AI server 200 according to an embodiment of the present disclosure.
[0020]
FIG. 2 illustrates an AI system 1 according to an embodiment of the present disclosure.
[0021]
FIG. 3 is a block diagram illustrating a configuration of an electronic device 300 accord…
