Artificial intelligence apparatus for recognizing speech including multiple languages, and method for the same
AssigneeLG Electronics Inc.
InventorsYejin Kim, Hyun Yu, Jonghoon Chae
Life of this patent
Estimating the term…
Abstract
An AI apparatus includes a microphone to acquire speech data including multiple languages, and a processor to acquire text data corresponding to the speech data, determine a main language from languages included in the text data, acquire a translated text data obtained by translating a text data portion, which has a language other than the main language, in the main language, acquire a morpheme analysis result for the translated text data, extract a keyword for intention analysis from the morpheme analysis result, acquire an intention pattern matched to the keyword, and perform an operation corresponding to the intention pattern.
Claims18
An artificial intelligence (AI) apparatus comprising:
- a microphone to acquire speech data including multiple languages;
- and
- a processor configured to:
- generate text data corresponding to the speech data;
- determine a primary language from the multiple languages included in the generated text data based on a determination of which language is first tagged sequentially in the generated text data;
- translate, to the primary language, portions of the generated text data that are in a language other than the primary language to obtain translated text data, wherein the obtained translated text data corresponds to the generated text data in the primary language;
- generate a morpheme analysis result based on the obtained translated text data;
- extract a keyword from the morpheme analysis result for determining an intention of the speech data;
- and
- cause an operation to be performed when an intention pattern is identified as corresponding to the extracted keyword, wherein the intention pattern is based on intention information extracted from various speech data.
A method for recognizing a speech including multiple languages, the method comprising:
- acquiring speech data including multiple languages;
- generating text data corresponding to the speech data;
- determining a primary language from the multiple languages included in the generated text data based on a determination of which language is first tagged sequentially in the generated text data;
- translating, to the primary language, portions of the generated text data that are in a language other than the primary language to obtain translated text data, wherein the obtained translated text data corresponds to the generated text data in the primary language;
- generating a morpheme analysis result based on the obtained translated text data;
- extracting a keyword from the morpheme analysis result for determining an intention of the speech data;
- and
- causing an operation to be performed when an intention pattern is identified as corresponding to the extracted keyword, wherein the intention pattern is based on intention information extracted from various speech data.
Dependencies are inferred from the claim text.