Methods, systems, and products for indexing content

• US 8,694,318 B2
• Filed: 09/19/2006
• Issued: 04/08/2014
• Est. Priority Date: 09/19/2006
• Status: Expired due to Fees

• Current Assignee: Bellsouth Intellectual Property Corporation (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Yu, Ke
• Primary Examiner(s): Godbold, Douglas
• Assistant Examiner(s): Villena, Mark
• Application Number:
  US11/523,413
• Publication Number:
  US 20080071542A1
• Time in Patent Office: 2,758 Days
• Field of Search: 704/254, 704/270, 704/267, 704/275, 704/260, 704/231, 352/5, 707/104.1, 707/102, 707/3
• US Class Current: 704/254
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units

Abstract

Methods, systems, and products index content. An audio stream is converted to a stream of phonemes. The stream of phonemes is filtered according to a list of popular phoneme strings. A weighting is computed for the content based on at least one popular phoneme combination occurring in the stream of phonemes. The content is indexed based on the weighting.

Claims

1. A method of indexing content, comprising:

receiving video content comprising an audio stream;

converting the audio stream to a string of phonemes;

converting each phoneme in the string of phonemes to an amplitude and a frequency component for each phonetic unit;

storing associations in a database between popular text search queries and corresponding popular amplitudes and frequencies;

inspecting amplitude and frequency components corresponding to the string of phonemes for each occurrence of a popular amplitude and frequency corresponding to a popular text query;

retrieving a score for each popular amplitude and frequency component found in the amplitude and frequency components corresponding to the string of phonemes;

summing each score of each occurrence of the popular amplitude and frequency component to obtain a total score for the audio stream;

storing associations in the database between titles of video content to total scores for audio streams;

adding an entry in the database that associates a title of the video content to the total score for the audio stream;

receiving a text query;

querying the database for the text query and a threshold score;

retrieving the titles of video content matching the text query and having the total score at least equal to the threshold score; and

responding to the text query with the titles of video content.

2. The method according to claim 1, further comprising accessing a database of popular search queries and retrieving a search string of phonemes corresponding to a popular search query.

3. The method according to claim 1, further comprising storing popular phoneme combinations occurring in the string of phonemes.

4. The method according to claim 3, further comprising:

associating the popular phoneme combinations occurring in the string of phonemes to popular words; and

indexing the popular words to the content.

5. The method according to claim 4, further comprising:

converting the text query to the string of phonemes; and

comparing the string of phonemes corresponding to the text query to an index of popular phoneme combinations.

6. A system of indexing content, comprising:

a processor;

memory; and

code stored in the memory that causes the processor at least to perform operations, the operations comprising;

receive video content comprising an audio stream;

convert the audio stream to a string of phonemes;

convert each phoneme in the string of phonemes to an amplitude and a frequency component for each phonetic unit;

store associations in a database between popular text search queries and corresponding popular amplitudes and frequencies;

inspect amplitude and frequency components corresponding to the string of phonemes for each occurrence of a popular amplitude and frequency corresponding to a popular text query;

retrieve a score for each popular amplitude and frequency component found in the amplitude and frequency components corresponding to the string of phonemes;

sum each score of each occurrence of the popular amplitude and frequency component to obtain a total score for the audio stream;

store associations in the database between titles of video content to total scores for audio streams;

add an entry in the database that associates a title of the video content to the total score for the audio stream;

receive a text query;

query the database for the text query and a threshold score;

retrieve the titles of video content matching the text query and having the total score at least equal to the threshold score; and

respond to the text query with the titles of video content.

7. The system according to claim 6, wherein the code further causes the processor to access a database of popular search queries and convert each popular search query to a search string of phonemes.

8. The system according to claim 6, wherein the code further causes the processor to store in the memory popular phoneme combinations occurring in the string of phonemes.

9. The system according to claim 6, wherein the code further causes the processor to:

associate popular phoneme combinations occurring in the string of phonemes to popular words; and

index the popular words to the content.

10. The system according to claim 9, wherein the code further causes the processor to:

compare the text query to an index of popular words.

11. The system according to claim 9, wherein the code further causes the processor to:

convert the text query to a search string of phonemes; and

compare the search string of phonemes to an index of popular phoneme combinations.

12. A non-transitory computer readable medium storing computer-readable instructions for performing a method, the method comprising:

receiving video content comprising an audio stream;

converting the audio stream to a string of phonemes;

converting each phoneme in the string of phonemes to an amplitude and a frequency component for each phonetic unit;

storing associations in a database between popular text search queries and corresponding popular amplitudes and frequencies;

inspecting amplitude and frequency components corresponding to the string of phonemes for each occurrence of a popular amplitude and frequency corresponding to a popular text query;

retrieving a score for each popular amplitude and frequency component found in the amplitude and frequency components corresponding to the string of phonemes;

summing each score of each occurrence of the popular amplitude and frequency component to obtain a total score for the audio stream;

storing associations in the database between titles of video content to total scores for audio streams;

adding an entry in the database that associates a title of the video content to the total score for the audio stream;

receiving a text query;

querying the database for the text query and a threshold score;

retrieving the titles of video content matching the text query and having the total score at least equal to the threshold score; and

responding to the text query with the titles of video content.

13. The computer readable medium according to claim 12, further comprising instructions for accessing a database of popular search queries and converting each popular search query to a search string of phonemes.

14. The computer readable medium according to claim 13, further comprising instructions for converting the search string of phonemes to amplitudes and frequencies.

15. The computer readable medium according to claim 12, further comprising instructions for storing in memory popular phoneme combinations.

16. The computer readable medium according to claim 15, further comprising instructions for:

associating the popular phoneme combinations to popular words; and

indexing the popular words to the content.

17. The computer readable medium according to claim 16, further comprising instructions for:

receiving a search query comprising text data; and

comparing the text query data to an index of popular words.