Generating query answers
A server receives a query identifying one or more attributes of an entity. The server accesses a set of candidate templates for answering the query based on the one or more attributes of the entity, each candidate template having one or more fields, wherein each field is associated with at least one constraint. The server obtains a set of information that answers the query and selects a template from the set of candidate templates, where the selected template has a maximum number of fields with constraints that are satisfied by the set of information. The server then generates a phrase by adding the set of information to the one or more fields of the selected template, such that the phrase comprises an answer to the query. Finally, the server communicates the phrase to a client device.
1. A non-transitory computer-readable medium storing instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
receiving, at a server, a query, wherein the query represents an utterance that is converted into a transcription;
responsive to receiving the query, performing a search and obtaining, at the server, a set of information that answers the query, wherein the set of information that answers the query comprises a set of entity-attribute-value triples;
after obtaining the set of information that answers the query, selecting, at the server and from among a set of candidate templates, a template from which to generate a phrase that comprises an answer to the query, the selected template providing a format for populating a number of eligible fields with the set of information, wherein an eligible field is a field having a constraint that is satisfied by the set of information and wherein the selected template is selected based on the number of eligible fields with constraints that are satisfied by the set of information;
generating, at the server, the phrase by populating the eligible fields of the selected template with the set of information, such that the phrase comprises an answer to the query; and
communicating, from the server, an audio signal corresponding to the phrase to a client device.
2. The non-transitory computer-readable medium of claim 1 , wherein the constraints comprise one or more of a type constraint, a temporal constraint, a gender constraint, a relationship constraint, a singular/plural constraint, a unit of measure constraint, and a determinant constraint.
3. The non-transitory computer-readable medium of claim 1 , wherein the phrase comprises a sentence.
4. The non-transitory computer-readable medium of claim 1 ,
wherein obtaining, at the server, a set of information that answers the query comprises obtaining at the server, a plurality of sets of information that answer the query;
wherein selecting, at the server, the template from which to generate the phrase that comprises an answer to the query comprises selecting at the server, for each of the plurality of sets of information, a template from the set of candidate templates;
wherein generating, at the server, the phrase by populating the eligible fields of the selected template with the set of information such that the phrase comprises an answer to the query comprises generating at the server, tor each of the selected templates, a phrase by populating the eligible fields of the respective selected template with the respective set of information; and
wherein communicating, from the server, the audio signal corresponding to the phrase for output by the client device comprises communicating, from the server, an audio signal corresponding to a sentence including the phrases to the client device,
wherein the operations further comprise,
obtaining, at the server, a sentence template based on a type of entity, wherein the sentence template includes a plurality of fields for the phrases; and
adding the phrases to the fields of the sentence template to form the sentence.
5. The non-transitory computer-readable medium of claim 1 , wherein:
the phrase comprises a sentence; and
communicating, from the server, the audio signal corresponding to the phrase to the client device comprises communicating, from the server, an audio signal corresponding to the sentence to the client device.
6. A computer-implemented method comprising:
receiving, at a server, a query, wherein the query represents an utterance that is converted into a transcription;
responsive to receiving the query, performing a search and obtaining, at the server, a set of information that answers the query, wherein the set of information that answers the query comprises a set of entity-attribute-value triples;
after obtaining the set of information that answers the query, selecting, at the server and from among a set of candidate templates, a template from which to generate a phrase that comprises an answer to the query, the selected template providing a format for populating a number of eligible fields with the set of information, wherein an eligible field is a field having a constraint that is satisfied by the set of information and wherein the selected template is selected based on the number of eligible fields with constraints that are satisfied by the set of information;
generating, at the server, the phrase by populating the eligible fields of the selected template with the set of information, such that the phrase comprises an answer to the query; and
communicating, from the server, an audio signal corresponding to the phrase to a client device.
7. The method of claim 6 , wherein the constraints comprise one or more of a type constraint, a temporal constraint, a gender constraint, a relationship constraint, a singular/plural constraint, a unit of measure constraint, and a determinant constraint.
8. The method of claim 6 , wherein the phrase comprises a sentence.
9. The method of claim 6 ,
wherein obtaining, at the server, a set of information that answers the query comprises obtaining at the server, a plurality of sets of information that answer the query;
wherein selecting, at the server, the template from which to generate the phrase that comprises an answer to the query comprises selecting at the server, for each of the plurality of sets of information, a template from the set of candidate templates;
wherein generating, at the server, the phrase by populating the eligible fields of the selected template with the set of information, such that the phrase comprises an answer to the query comprises generating at the server, for each of the selected templates, a phrase by populating the eligible fields of the respective selected template with the respective set of information; and
wherein communicating, from the server, the audio signal corresponding to the phrase for output by the client device comprises communicating, from the server, an audio signal corresponding to a sentence including the phrases to the client device,
further comprising,
obtaining, at the server, a sentence template based on a type of entity, wherein the sentence template includes a plurality of fields for the phrases; and
adding the phrases to the fields of the sentence template to form the sentence.
10. The method of claim 6 , wherein:
the phrase comprises a sentence; and
communicating, from the server, the audio signal corresponding to the phrase to the client device comprises communicating, from the server, an audio signal corresponding to the sentence to the client device.
11. 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 a query, wherein the query represents an utterance that is converted into a transcription;
responsive to receiving the query, performing a search and obtaining a set of information that answers the query, wherein the set of information that answers the query comprises a set of entity-attribute-value triples;
after obtaining the set of information that answers the query, selecting, at the one or more computers and from among a set of candidate templates, a template from which to generate a phrase that comprises an answer to the query, the selected template providing a format for populating a number of eligible fields with the set of information, wherein an eligible field is a field having a constraint that is satisfied by the set of information and wherein the selected template is selected based on the number of eligible fields with constraints that are satisfied by the set of information;
generating the phrase by populating the eligible fields of the selected template with the set of information, such that the phrase comprises an answer to the query; and
communicating an audio signal corresponding to the phrase to a client device.
12. The system of claim 11 , wherein the constraints comprise one or more of a type constraint, a temporal constraint, a gender constraint, a relationship constraint, a singular/plural constraint, a unit of measure constraint, and a determinant constraint.
13. The system of claim 11 , wherein the phrase comprises a sentence.
14. The system of claim 11 ,
wherein obtaining a set of information that answers the query comprises obtaining at the server, a plurality of sets of information that answer the query;
wherein selecting the template from which to generate the phrase that comprises an answer to the query comprises selecting at the server, for each of the plurality of sets of information, a template from the set of candidate templates;
wherein generating the phrase by populating the eligible fields of the selected template with the set of information, such that the phrase comprises an answer to the query comprises generating at the server, for each of the selected templates, a phrase by populating the eligible fields of the respective selected template with the respective set of information; and
wherein communicating the audio signal corresponding to the phrase for output by the client device comprises communicating an audio signal corresponding to a sentence including the phrases to the client device,
wherein the operations further comprise,
obtaining a sentence template based on a type of entity, wherein the sentence template includes a plurality of fields for the phrases; and
adding the phrases to the fields of the sentence template to form the sentence.
15. The system of claim 11 , wherein:
the phrase comprises a sentence; and
communicating the audio signal corresponding to the phrase to the client device comprises communicating an audio signal corresponding to the sentence to the client device.