Method for searching for texts in different languages based on pronunciation and electronic device applying the same

A method of searching for a text in different languages based on pronunciation, performed by an electronic device, the method comprising:

• collecting a first input text string based on data stored in the electronic device;
• converting the first input text string into a first standard text string based on language information set in the electronic device and a standard conversion table including a plurality of first texts generated based on a romanization rule or an international phonetic alphabet (IPA), wherein the converted first standard text string includes a plurality of second texts corresponding to the first input text string among the plurality of first texts;
• converting the first standard text string into a first code string based on a code translation table including a plurality of first codes corresponding to each of a plurality of groups in which texts having similar pronunciations among the plurality of first texts are classified, wherein the first code string includes a plurality of second codes corresponding to each of the plurality of second texts among the plurality of the first codes;
• generating and storing a plurality of index vectors based on two consecutive codes included in the first code string and an order of each of the two consecutive codes included in the first code string;
• further comprising:
• if a query including a second input text string which is written in a first language is received, converting the second input text string into a second standard text string based on the language information set in the electronic device and the standard conversion table, wherein the converted second standard text string includes a plurality of third texts corresponding to the second input text string among the plurality of first texts;
• converting the second standard text string into a second code string based on the code conversion table, wherein the second code string includes a plurality of third codes corresponding to each of the plurality of third texts among the plurality of the first codes;
• generating a plurality of first vectors based on two consecutive codes included in the second code string and an order of each of the two consecutive codes included in the second code string;
• comparing the plurality of first vectors with the plurality of index vectors stored in the electronic device;
• identifying a plurality of second vectors having a similarity, which is greater than or equal to a threshold, with the plurality of first vectors from among the plurality of index vectors based on the comparison;
• and
• providing, as a response to the query, an output text string corresponding to the plurality of second vectors,
• wherein the output text string is written in a second language different from the first language, and
• wherein pronunciation of the second input text string corresponds to pronunciation of the output text string.

A computer program product stored on a non-transitory machine-readable storage medium comprising instructions that, when executed, cause a processor of an electronic device to:

• collect a first input text string based on data stored in the electronic device;
• convert the first input text string into a first standard text string based on language information set in the electronic device and a standard conversion table including a plurality of first texts generated based on a romanization rule or an international phonetic alphabet (IPA), wherein the converted first standard text string includes a plurality of second texts corresponding to the first input text string among the plurality of first texts;
• convert the first standard text string into a first code string based on a code conversion table including a plurality of first codes corresponding to each of a plurality of groups in which texts having similar pronunciations among the plurality of first texts are classified, wherein the first code string includes a plurality of second codes corresponding to each of the plurality of second texts among the plurality of the first codes;
• generate and storing a plurality of index vectors based on two consecutive codes included in the first code string and an order of each of the two consecutive codes included in the first code string;
• wherein the instructions that, when executed, further cause the processor of an electronic device to:
• if a query including a second input text string which is written in a first language is received, convert the second input text string into a second standard text string based on the language information set in the electronic device and the standard conversion table, wherein the converted second standard text string includes a plurality of third texts corresponding to the second input text string among the plurality of first texts;
• convert the second standard text string into a second code string based on the code conversion table, wherein the second code string includes a plurality of third codes corresponding to each of the plurality of third texts among the plurality of the first codes;
• generate a plurality of first vectors based on two consecutive codes included in the second code string and an order of each of the two consecutive codes included in the second code string;
• compare the plurality of first vectors with the plurality of index vectors stored in the electronic device;
• identify a plurality of second vectors having a similarity, which is greater than or equal to a threshold, with the plurality of first vectors from among the plurality of index vectors based on the comparison;
• and
• provide, as a response to the query, an output text string corresponding to the plurality of second vectors,
• wherein the output text string is written in a second language different from the first language, and
• wherein pronunciation of the second input text string corresponds to pronunciation of the output text string.