17432446. ELECTRONIC APPARATUS AND METHOD FOR CONTROLLING THEREOF simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

ELECTRONIC APPARATUS AND METHOD FOR CONTROLLING THEREOF

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Hyojung Han of Suwon-si Gyeonggi-do (KR)

Sathish Indurthi of Suwon-si Gyeonggi-do (KR)

Beomseok Lee of Suwon-si Gyeonggi-do (KR)

Mohd Abbas Zaidi of Suwon-si Gyeonggi-do (KR)

Nikhil Kumar of Suwon-si Gyeonggi-do (KR)

ELECTRONIC APPARATUS AND METHOD FOR CONTROLLING THEREOF - A simplified explanation of the abstract

This abstract first appeared for US patent application 17432446 titled 'ELECTRONIC APPARATUS AND METHOD FOR CONTROLLING THEREOF

Simplified Explanation

The abstract describes an electronic apparatus that uses a neural network model to translate sentences from one language to another. Here are the key points:

  • The electronic apparatus includes an inputter, memory, and processor.
  • The inputter acquires an input sentence in a first language, consisting of multiple tokens.
  • The memory stores instructions related to the translation process.
  • The processor controls the apparatus to perform various operations for translation.
  • At a first time point, the processor acquires a subset of input tokens and translates them using a neural network model.
  • At a second time point, the processor identifies an operation to be performed, either identifying additional input tokens or generating output tokens.
  • The processor acquires a number corresponding to the identified operation.
  • Finally, the processor translates the input sentence into a second language based on the identified operation and the acquired number.

Potential applications of this technology:

  • Language translation: The electronic apparatus can be used in applications that require real-time translation between different languages.
  • Communication devices: It can be integrated into devices like smartphones or tablets to provide instant translation capabilities.
  • Language learning: The apparatus can assist language learners by providing accurate translations and helping them understand the structure of different languages.

Problems solved by this technology:

  • Language barrier: The apparatus helps overcome language barriers by providing quick and accurate translations.
  • Efficiency: By using a neural network model, the translation process can be performed efficiently, allowing for real-time translation.

Benefits of this technology:

  • Real-time translation: The electronic apparatus enables instant translation, making communication between people speaking different languages easier.
  • Accuracy: By utilizing a neural network model, the translation process can achieve high levels of accuracy.
  • Adaptability: The apparatus can be trained on different language pairs, making it adaptable to various translation needs.


Original Abstract Submitted

An electronic apparatus is disclosed. The electronic apparatus includes: an inputter comprising input circuitry configured to acquire an input sentence of a first language including a plurality of input tokens, a memory configured to store at least one instruction, and a processor, wherein the processor is configured to control the electronic apparatus to: acquire at least one first input token identified up to a first time point among the plurality of input tokens and acquire at least one first output token generated by translating the at least one first input token by inputting the at least one first input token and the at least one first output token into a learned first neural network model, identify an operation to be performed at a second time point after the first time point among a first operation configured to identify at least one second input token other than the at least one first input token among the plurality of input tokens and a second operation configured to generate at least one second output token based on the at least one first input token, and acquire a number corresponding to the identified operation and acquiring an output sentence of a second language different from the first language by translating the input sentence based on the identified operation and the acquired number.