Electronic device and speech processing method thereof
According to various example embodiments, an electronic device includes a microphone configured to receive an audio signal including speech of a user, a processor, and a memory configured to store instructions executable by the processor and personal information of the user, in which the processor is configured to extract a plurality of speech recognition candidates by analyzing a feature of the speech of the user, extract a keyword based on the plurality of speech recognition candidates, search for replacement data, based on the keyword and the personal information, and generate a recognition result corresponding to the speech of the user, based on the replacement data.
1. An electronic device comprising:
a microphone configured to receive an audio signal including speech of a user,
a processor, and
a memory configured to store instructions executable by the processor and personal information of the user,
wherein the processor is configured to:
extract a plurality of speech recognition candidates by analyzing a feature of the speech of the user,
extract a keyword based on the plurality of speech recognition candidates,
search for replacement data to replace the keyword, based on the keyword and the personal information,
generate a recognition result corresponding to the speech of the user, based on the replacement data, and
perform security processing on the recognition result, based on the personal information.
2. The electronic device of claim 1 , wherein
the processor is configured to:
extract a feature vector based on the speech of the user, and
extract the plurality of speech recognition candidates based on the feature vector.
3. The electronic device of claim 2 , wherein
the processor is configured to:
calculate probability values respectively corresponding to the plurality of speech recognition candidates, based on the feature vector, and
determine one or more target candidates based on the probability values.
4. The electronic device of claim 1 , wherein
the processor is configured to:
calculate a first probability value by inputting the plurality of speech recognition candidates to a first language model, which models a general feature of a language,
calculate a second probability value by inputting the plurality of speech recognition candidates to a second language model, which models a linguistic feature of the user, and
determine one or more target candidates, based on the first probability value and the second probability value.
5. The electronic device of claim 1 , wherein
the processor is configured to extract the keyword, based on a database related to the plurality of speech recognition candidates and to the speech of the user.
6. The electronic device of claim 5 , wherein
the database related to the speech of the user comprises a feature of utterance of the user, a feature of vocalization of the user, a feature of pronunciation of the user, a sentence structure included in the speech of the user, or grammar of a sentence included in the speech of the user.
7. The electronic device of claim 5 , wherein
the processor is configured to:
detect an entity name from the recognition result, based on the database related to the speech of the user,
obtain a category corresponding to the entity name, based on a preset rule, and search for the replacement data based on the category.
8. The electronic device of claim 7 , wherein
the entity name comprises:
a person name included in the speech of the user, a name of an application related to the speech of the user, or a name of content included in the speech of the user, and
the category comprises:
a schedule included in the speech of the user or an operation related to the speech of the user.
9. The electronic device of claim 7 , wherein
the processor is configured to:
search for similar data to the category from the personal information,
calculate a similarity between the similar data and the entity name, and
determine the similar data to be the replacement data, based on the similarity.
10. An electronic device comprising:
a microphone configured to receive an audio signal including speech of a user;
a processor; and
a memory configured to store instructions executable by the processor and personal information of the user,
wherein
the processor is configured to:
extract a plurality of speech recognition candidates by analyzing a feature of the speech of the user,
extract a keyword based on the plurality of speech recognition candidates,
search for replacement data to replace the keyword, based on the keyword and the personal information,
generate a recognition result corresponding to the speech of the user, based on the replacement data, and
perform security processing on the recognition result, based on the personal information.
11. The electronic device of claim 10 , wherein
the processor is configured to:
extract a feature vector based on the speech of the user, and
extract the plurality of speech recognition candidates based on the feature vector.
12. The electronic device of claim 11 , wherein
the processor is configured to:
calculate probability values respectively corresponding to the plurality of speech recognition candidates, based on the feature vector, and
determine one or more target candidates based on the probability values.
13. The electronic device of claim 10 , wherein
the processor is configured to:
calculate a first probability value by inputting the plurality of speech recognition candidates to a first language model, which models a general feature of a language,
calculate a second probability value by inputting the plurality of speech recognition candidates to a second language model, which models a linguistic feature of the user, and
determine one or more target candidates, based on the first probability value and the second probability value.
14. The electronic device of claim 10 , wherein
the processor is configured to extract the keyword, based on a database related to the plurality of speech recognition candidates and to the speech of the user.
15. The electronic device of claim 14 , wherein
the database related to the speech of the user comprises a feature of utterance of the user, a feature of vocalization of the user, a feature of pronunciation of the user, a sentence structure included in the speech of the user, or grammar of a sentence included in the speech of the user.
16. The electronic device of claim 15 , wherein
the processor is configured to:
detect an entity name from the recognition result, based on the database related to the speech of the user,
obtain a category corresponding to the entity name, based on a preset rule, and
search for the replacement data based on the category.
17. The electronic device of claim 16 , wherein
the entity name comprises:
a person name included in the speech of the user, a name of an application related to the speech of the user, or a name of content included in the speech of the user, and
the category comprises:
a schedule included in the speech of the user or an operation related to the speech of the user.
18. The electronic device of claim 16 , wherein
the processor is configured to:
search for similar data to the category from the personal information,
calculate a similarity between the similar data and the entity name, and
determine the similar data to be the replacement data, based on the similarity.
19. The electronic device of claim 10 , wherein
the processor is configured to:
determine whether a schedule of the user and a position of the user are included in the recognition result, based on the personal information, and
perform security processing by performing masking or deletion on the recognition result, based on a determination result.
20. A speech processing method of an electronic device, the method comprising:
receiving an audio signal including speech of a user using a microphone,
extracting a plurality of speech recognition candidates by analyzing a feature of the speech of the user,
extracting a keyword based on the plurality of speech recognition candidates;
searching for replacement data to replace the keyword, based on the keyword and personal information of the user,
generating a recognition result corresponding to the speech of the user, based on the replacement data, and
performing security processing on the recognition result, based on the personal information.