Background audio identification for query disambiguation
Implementations relate to techniques for providing context-dependent search results. The techniques can include receiving a query and background audio. The techniques can also include identifying the background audio, establishing concepts related to the background audio and obtaining terms related to the concepts related to the background audio. The techniques can also include obtaining search results based on the query and on at least one of the terms. The techniques can also include providing the search results.
1. A computer-implemented method comprising:
receiving (i) a search query including one or more query terms entered into a mobile computing device by a user, the search query being entered by the user speaking the search query during a first period of time during which a voice detection signal is above a threshold volume level, indicating the presence of the user's voice, and (ii) background audio that is not made by the user, and that is produced by a source in an environment surrounding the mobile computing device within a predetermined time of entry of the one or more query terms, the predetermined time of entry being outside of the first time period, and wherein the background audio includes audio that is detected during a second time period that is a fixed time interval and after the first time period during which the voice detection signal is below the threshold volume level, indicating an absence of the user's voice, the background audio being detected during the second time period in response to a determination by the mobile computing device that the voice detection signal has fallen below the threshold volume level;
identifying a known audio segment based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
generating a set of related terms that describe entities that are associated in an entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
modifying the search query to include, as query terms, one or more related terms of the set of related terms that describe entities that are associated in the entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
receiving search results based on the modified search query; and
providing the search results.
2. The method of claim 1 , wherein receiving search results based on the modified search query comprises:
providing the query to a search engine;
receiving scored results from the search engine; and
altering a score for a search result containing at least one of the related terms.
3. The method of claim 2 , further comprising determining an amount to alter the score by using a plurality of training queries.
4. The method of claim 1 , wherein providing the search results comprises providing only search results that satisfy a threshold score.
5. The method of claim 1 , wherein identifying the known audio segment based on the background audio comprises:
recognizing at least a portion of the background audio by matching it to an acoustic fingerprint; and
identifying the known audio segment based on the background audio comprising the known audio segment associated with the acoustic fingerprint.
6. The method of claim 1 , wherein generating the set of terms related to the background audio comprises:
generating the set of terms based on querying a database based on the known audio segment based on the background audio.
7. A system comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
receiving (i) a search query including one or more query terms entered into a mobile computing device by a user, the search query being entered by the user speaking the search query during a first period of time during which a voice detection signal is above a threshold volume level, indicating the presence of the user's voice, and (ii) background audio that is not made by the user, and that is produced by a source in an environment surrounding the mobile computing device within a predetermined time of entry of the one or more query terms, the predetermined time of entry being outside of the first time period, and wherein the background audio includes audio that is detected during a second time period that is a fixed time interval and after the first time period during which the voice detection signal is below the threshold volume level, indicating an absence of the user's voice, the background audio being detected during the second time period in response to a determination by the mobile computing device that the voice detection signal has fallen below the threshold volume level;
identifying a known audio segment based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
generating a set of related terms that describe entities that are associated in an entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
modifying the search query to include, as query terms, one or more related terms of the set of related terms that describe entities that are associated in the entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
receiving search results based on the modified search query; and
providing the search results.
8. The system of claim 7 , wherein receiving search results based on the modified search query comprises:
providing the query to a search engine;
receiving scored results from the search engine; and
altering a score for a search result containing at least one of the related terms.
9. System of claim 8 , wherein the operations further comprise determining an amount to alter the score by using a plurality of training queries.
10. The system of claim 7 , wherein providing the search results comprises providing only search results that satisfy a threshold score.
11. The system of claim 7 , wherein identifying the known audio segment based on the background audio comprises:
recognizing at least a portion of the background audio by matching it to an acoustic fingerprint; and
identifying the known audio segment based on the background audio comprising the known audio segment associated with the acoustic fingerprint.
12. The system of claim 7 , wherein generating the set of terms related to the background audio comprises:
generating the set of terms based on querying a database based on the known audio segment based on the background audio.
13. A non-transitory computer-readable storage device storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
receiving (i) a search query including one or more query terms entered into a mobile computing device by a user, the search query being entered by the user speaking the search query during a first period of time during which a voice detection signal is above a threshold volume level, indicating the presence of the user's voice, and (ii) background audio that is not made by the user, and that is produced by a source in an environment surrounding the mobile computing device within a predetermined time of entry of the one or more query terms, the predetermined time of entry being outside of the first time period, and wherein the background audio includes audio that is detected during a second time period that is a fixed time interval and after the first time period during which the voice detection signal is below the threshold volume level, indicating an absence of the user's voice, the background audio being detected during the second time period in response to a determination by the mobile computing device that the voice detection signal has fallen below the threshold volume level;
identifying a known audio segment based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
generating a set of related terms that describe entities that are associated in an entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
modifying the search query to include, as query terms, one or more related terms of the set of related terms that describe entities that are associated in the entity-relationship model with the known audio segment that is identified based on the background audio that is not made by the user, and that is produced by the source in the environment surrounding the mobile computing device within the predetermined time of the entry of the one or more query terms;
receiving search results based on the modified search query; and
providing the search results.
14. The storage device of claim 13 , wherein receiving search results based on the modified search query comprises:
providing the query to a search engine;
receiving scored results from the search engine; and
altering a score for a search result containing at least one of the related terms.
15. The storage device of claim 14 , wherein the operations further comprise determining an amount to alter the score by using a plurality of training queries.
16. The storage device of claim 13 , wherein providing the search results comprises providing only search results that satisfy a threshold score.
17. The storage device of claim 13 , wherein identifying the known audio segment based on the background audio comprises:
recognizing at least a portion of the background audio by matching it to an acoustic fingerprint; and
identifying the known audio segment based on the background audio comprising the known audio segment associated with the acoustic fingerprint.
18. The storage device of claim 13 wherein generating the set of terms related to the background audio comprises:
generating the set of terms based on querying a database based on the known audio segment based on the background audio.