Conversational agent response determined using a sentiment
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for handing off a user conversation between computer-implemented agents. One of the methods includes receiving, by a computer-implemented agent specific to a user device, a digital representation of speech encoding an utterance, determining, by the computer-implemented agent, that the utterance specifies a requirement to establish a communication with another computer-implemented agent, and establishing, by the computer-implemented agent, a communication between the other computer-implemented agent and the user device.
1. A computer-implemented method comprising:
receiving, by a first computer-implemented agent for a user device, a text representation of an utterance that includes a command, wherein the text representation of the utterance is determined based on a speech encoding of the utterance; and
in response to processing the text representation of the utterance to determine words included in the utterance:
determining, from among a plurality of different demographics, a particular demographic of a speaker of the utterance;
in response to determining the particular demographic, selecting, from a plurality of different computer-implemented agents, a particular computer- implemented agent based on the particular demographic associated with the particular computer-implemented agent matching the particular demographic, wherein each agent is associated with a respective one of the plurality of different demographics; and
causing the particular computer-implemented agent to provide an interface for processing subsequent utterances spoken by the speaker.
2. The computer-implemented method of claim 1 , wherein the command comprises a request for results responsive to the first computer-implemented agent that is different from the particular computer-implemented agent.
3. The computer-implemented method of claim 1 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic using the text representation of the utterance.
4. The computer-implemented method of claim 1 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on profile data that includes demographic information for the speaker.
5. The computer-implemented method of claim 3 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on a style of speech determined from the speech encoding of the utterance.
6. The computer-implemented method of claim 3 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises receiving data specifying the particular demographic based on a response from the speaker.
7. A user device comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, are operable to:
receive, by a first computer-implemented agent for the user device, a text representation of an utterance that includes a command, wherein the text representation of the utterance is determined based on a speech encoding of the utterance; and
in response to processing the text representation of the utterance to determine words included in the utterance:
determine, from among a plurality of different demographics, a particular demographic of a speaker of the utterance;
in response to determining the particular demographic, select, from a plurality of different computer-implemented agents, a particular computer-implemented agent based on the particular demographic associated with the particular computer-implemented agent matching the particular demographic, wherein each agent is associated with a respective one of the plurality of different demographics; and
cause the particular computer-implemented agent to provide an interface for processing subsequent utterances spoken by the speaker.
8. The user device of claim 7 , wherein the command comprises a request for results responsive to the first computer-implemented agent that is different from the particular computer-implemented agent.
9. The user device of claim 7 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic using the text representation of the utterance.
10. The user device of claim 7 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on profile data that includes demographic information for the speaker.
11. The user device of claim 9 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on a style of speech determined from the speech encoding of the utterance.
12. The user device of claim 9 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises receiving data specifying the particular demographic based on a response from the speaker.
13. A non-transitory computer readable storage medium storing instructions executable by a data processing apparatus and, upon such execution, causes the data processing apparatus to perform operations comprising:
receiving, by a first computer-implemented agent for a user device, a text representation of an utterance that includes a command, wherein the text representation of the utterance is determined based on a speech encoding of the utterance; and
in response to processing the text representation of the utterance to determine words included in the utterance:
determining, from among a plurality of different demographics, a particular demographic of a speaker of the utterance;
in response to determining the particular demographic, selecting, from a plurality of different computer-implemented agents, a particular computer-implemented agent based on the particular demographic associated with the particular computer-implemented agent matching the particular demographic, wherein each agent is associated with a respective one of the plurality of different demographics; and
cause the particular computer-implemented agent to provide an interface for processing subsequent utterances spoken by the speaker.
14. The non-transitory computer readable storage medium of claim 13 , wherein the command comprises a request for results responsive to the first computer-implemented agent that is different from the particular computer-implemented agent.
15. The non-transitory computer readable storage medium of claim 13 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic using the text representation of the utterance.
16. The non-transitory computer readable storage medium of claim 13 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on profile data that includes demographic information for the speaker.
17. The non-transitory computer readable storage medium of claim 16 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises determining the particular demographic based on a style of speech determined from the speech encoding of the utterance.
18. The non-transitory computer readable storage medium of claim 16 , wherein determining, from among the plurality of different demographics, the particular demographic of the speaker of the utterance comprises receiving data specifying the particular demographic based on a response from the speaker.