Machine translation method and apparatus

Patent metadata

Field	Value
Publication number	US-10067939-B2
Application number	US-201715401126-A
Country	US
Kind code	B2
Filing date	Jan 9, 2017
Priority date	Aug 16, 2016
Publication date	Sep 4, 2018
Grant date	Sep 4, 2018

Abstract

A machine translation method includes converting a source sentence written in a first language to language-independent information using an encoder for the first language, and converting the language-independent information to a target sentence corresponding to the source sentence and written in a second language different from the first language using a decoder for the second language. The encoder for the first language is trained to output language-independent information corresponding to the target sentence in response to an input of the source sentence.

First claim

What is claimed is:

1. A machine translation method comprising:
converting, using a processor, a source sentence written in a first language to language-independent information using an encoder for the first language, information for which is stored in non-volatile and volatile memory; and
converting, using the processor, the language-independent information to a target sentence corresponding to the source sentence and written in a second language different from the first language using a decoder for the second language, information for which is stored in the nonvolatile and volatile memory, in response to a similarity being higher than a preset threshold, wherein the similarity is determined from a comparison between pixel values of the language-independent information of the source sentence and pixel values of the language-independent information of the target sentence;
wherein the encoder for the first language is trained to output language-independent information corresponding to the target sentence in response to an input of the source sentence, using the processor, wherein the processor communicates with the nonvolatile and volatile memory via a bus.

2. The machine translation method of claim 1 , wherein the converting of the source sentence to the language-independent information comprises converting the source sentence to language-independent information having a similarity to the language-independent information corresponding to the target sentence higher than a preset threshold.

3. The machine translation method of claim 1 , wherein the encoder for the first language is further trained to encode a first training sentence included in parallel corpus data in the first language and the second language to output training language-independent information; and
the training language-independent information corresponds to information output by encoding a second training sentence included in the parallel corpus data in an encoder for the second language, the second training sentence corresponding to the first training sentence.

4. The machine translation method of claim 3 , wherein the decoder for the second language is trained to output the second training sentence in response to an input of the training language-independent information from the encoder for the first language.

5. The machine translation method of claim 1 , wherein the language-independent information to which the source sentence is converted comprises any one or any combination of audio, video, brainwaves, and images corresponding to the source sentence.

6. The machine translation method of claim 1 , wherein the encoder for the first language and the decoder for the second language are implemented by one or more processors.

7. A non-transitory computer-readable storage medium storing instructions that, when executed by the processor, cause the processor to perform the method of claim 1 .

8. A machine translation method comprising:
converting, using a processor, a source sentence written in a first language to first language-independent information associated with the first language using an encoder for the first language, information for which is stored in nonvolatile and volatile memory;
converting, using the processor, the first language-independent information to second language-independent information associated with a second language different from the first language using a decoder for language-independent information, information for which is stored in the nonvolatile and volatile memory, in response to a similarity being higher than a preset threshold, wherein the similarity is determined from a comparison between pixel values of the language-independent information of the source sentence and pixel values of the language-independent information of the target sentence; and
converting the second language-independent information to a target sentence corresponding to the source sentence and written in the second language using a decoder for the second language, using the processor, wherein the processor communicates with the nonvolatile and volatile memory via a bus.

9. The machine translation method of claim 8 , wherein the second language-independent information corresponds to language-independent information output by encoding the target sentence in an encoder for the second language.

10. The machine translation method of claim 8 , wherein the decoder for language-independent information is trained to output second training language-independent information associated with the second language in response to an input of first training language-independent information associated with the first language, the second training language-independent information corresponding to the first training language-independent information;
the first training language-independent information is language-independent information output by encoding a first training sentence included in parallel corpus data in the first language and the second language in the encoder for the first language; and
the second training language-independent information is language-independent information output by encoding a second training sentence included in the parallel corpus data in an encoder for the second language, the second training sentence corresponding to the first training sentence.

11. The machine translation method of claim 8 , wherein the first language-independent information comprises a set, a permutation or a structured form of any one or any combination of any two or more of audio, video, brainwaves, and images corresponding to the source sentence.

12. The machine translation method of claim 8 , wherein the encoder for the first language, the decoder for language-independent information, and the decoder for the second language are implemented by one or more processors.

13. A machine translation apparatus comprising:
a nonvolatile and volatile memory storing information to implement an encoder for a first language and a decoder for a second language different from the first language; and
a processor configured to communicate with the nonvolatile and volatile memory via a bus, implement the encoder for the first language and the decoder for the second language based on the information stored in the memory, and translate a source sentence written in the first language into a target sentence written in the second language using the encoder for the first language and the decoder for the second language;
wherein the processor is further configured to convert the source sentence to language-independent information using the encoder for the first language, and convert the language-independent information to the target sentence using the decoder for the second language, in response to a similarity being higher than a preset threshold, wherein the similarity is determined from a comparison between pixel values of the language-independent information of the source sentence and pixel values of the language-independent information of the target sentence; and
the encoder for the first language is trained, using the processor, to output language-independent information corresponding to the target sentence in response to an input of the source sentence.

14. The machine translation apparatus of claim 13 , wherein the processor is further configured to convert the source sentence to language-independent information having a similarity to the language-independent information corresponding to the target sentence higher than a preset threshold.

15. The machine translation apparatus of claim 13 , wherein the encoder for the first language is further trained to encode a first training sent

Assignees

• Samsung Electronics Co Ltd

Inventors

• NA Hwidong
• Lee Hoshik
• Choi Young Sang

Classifications

• G06F40/42 Primary

  Data-driven translation · CPC title

• G06F40/58 Primary

  Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation · CPC title

• G10L25/30

  using neural networks · CPC title

• G06F17/289 Primary

  Physics · mapped topic