최소 단어 이상 선택하여야 합니다.
최대 10 단어까지만 선택 가능합니다.
다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
NTIS 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
DataON 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Edison 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Kafe 바로가기국가/구분 | United States(US) Patent 등록 |
---|---|
국제특허분류(IPC7판) |
|
출원번호 | US-0240420 (2008-09-29) |
등록번호 | US-8583418 (2013-11-12) |
발명자 / 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 | 피인용 횟수 : 8 인용 특허 : 413 |
Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech i
Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.
1. A method for determining a native language of a text string associated with metadata of a media asset, the method comprising: at an electronic device comprising a processor and memory storing instructions for execution by the processor: undergoing one or more N-gram analyses at a word level to de
1. A method for determining a native language of a text string associated with metadata of a media asset, the method comprising: at an electronic device comprising a processor and memory storing instructions for execution by the processor: undergoing one or more N-gram analyses at a word level to determine a plurality of probabilities of occurrence of the text string, where each of the probabilities of occurrence correspond to a probability of occurrence of the text string in a particular language of a plurality of languages, wherein, for each language, the one or more N-gram analyses are based on a first set of probabilities of occurrence of words if the text string corresponds to a first type of metadata field associated with the media asset, and are based on a second set of probabilities of occurrence of words if the text string corresponds to a second type of metadata field associated with the media asset; anddetermining that the native language of the text string is a language that is associated with the highest probability of occurrence out of the plurality of probabilities of occurrence. 2. The method of claim 1 wherein the one or more N-gram analyses at a word level comprises: for each group of a number N of words in the text string, retrieving a plurality of probabilities, each of which corresponds to a particular language and represents the probability of occurrence of that group of N words in that particular language; andfor each language, calculating a total sum of the retrieved probabilities. 3. The method of claim 2 wherein determining the native language of the text string comprises determining that the native language is a language having the highest calculated total sum. 4. The method of claim 1 wherein the one or more N-gram analyses at a word level comprises a unigram analysis wherein, for each word in the text string, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that word in that particular language. 5. The method of claim 1 wherein the one or more N-gram analyses at a word level comprises a bigram analysis wherein, for each group of two adjacent words in the text string, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that group of words in that particular language. 6. The method of claim 1 wherein the one or more N-gram analyses at a word level comprises a trigram analysis wherein, for each group of three adjacent words in the text string, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that group of words in that particular language. 7. The method of claim 1 wherein the one or more N-gram analyses at a word level comprises any combination of a unigram analysis, a bigram analysis and a trigram analysis, wherein total probability sums are calculated under each such analysis and are weighted differently. 8. The method of claim 1 further comprising separating the text string into distinct words. 9. The method of claim 1 further comprising determining whether each word in the text string is in vocabulary by consulting a table that includes a list of words that are known in all known languages. 10. The method of claim 9 further comprising, for each word that is not in vocabulary, undergoing one or more N-gram analyses at a character level to determine a plurality of probabilities of occurrence of the word, where each of the probabilities of occurrence of the word correspond to a probability of occurrence of the word in a particular language of the plurality of languages. 11. The method of claim 10 wherein the one or more N-gram analyses at a character level comprises: for each group of a number N of characters in the word that is not in vocabulary, retrieving a plurality of probabilities, each of which corresponds to a particular language and represents the probability of occurrence of that group of N characters in that particular language; andfor each language, calculating a total sum of the retrieved probabilities. 12. The method of claim 10 wherein the one or more N-gram analyses at a character level comprises a unigram analysis wherein, for each character in the word that is not in vocabulary, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that character in that particular language. 13. The method of claim 10 wherein the one or more N-gram analyses at a character level comprises a bigram analysis wherein, for each group of two adjacent characters in the word that is not in vocabulary, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that group of characters in that particular language. 14. The method of claim 10 wherein the one or more N-gram analyses at a character level comprises a trigram analysis wherein, for each group of three adjacent characters in the word that is not in vocabulary, a plurality of probabilities are retrieved, each of which corresponds to a particular language and represents the probability of occurrence of that group of characters in that particular language. 15. The method of claim 10 wherein the one or more N-gram analyses at a character level comprises any combination of a unigram analysis, a bigram analysis and a trigram analysis, wherein total probability sums are calculated under each such analysis and are weighted differently. 16. The method of claim 10, wherein for each language, the one or more N-gram analyses at a character level are based on a first set of probabilities of occurrence of characters if the text string corresponds to the first metadata field associated with the media asset, and are based on a second set of probabilities of occurrence of characters if the text string corresponds to the second metadata field associated with the media asset. 17. The method of claim 1, wherein the first metadata field associated with the media asset is a title of a media asset. 18. The method of claim 17, wherein the second metadata field associated with the media asset is any of an artist, a performer, or a composer of a media asset. 19. The method of claim 1, wherein the media asset is an audio file. 20. The method of claim 1, wherein the first metadata field corresponds to a first category of metadata, and the second metadata field corresponds to a second category of metadata.
Copyright KISTI. All Rights Reserved.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.