Techniques for concurrent processing of user speech
A server receives a user audio stream, the stream comprising multiple utterances. A query-processing module of the server continuously listens to and processes the utterances. The processing includes parsing successive utterances and recognizing corresponding queries, taking appropriate actions while the utterances are being received. In some embodiments, a query may be parsed and executed before the previous query's execution is complete.
1. A computer-implemented method of generating a response to a spoken input, the computer-implemented method comprising:
obtaining an audio input stream;
detecting in the audio input stream a beginning of a first utterance;
detecting in the audio input stream an end of the first utterance;
responsive to detecting the end of the first utterance:
starting to process the first utterance to recognize a first query;
starting to execute the first query to determine a first response;
continuing to process the audio input stream; and
detecting in the audio input stream a beginning of a second utterance;
responsive to detecting the end of the second utterance:
starting to process the second utterance to recognize a second query;
determining that there is not a serial constraint between the first query and the second query; and
responsive to determining that there is not a serial constraint between the first query and the second query, starting execution of the second query before the end of the execution of first query.
2. The computer-implemented method of claim 1 , further comprising outputting the second response before the first response.
3. The computer-implemented method of claim 2 , wherein outputting the second response is performed concurrently with executing the first query.
4. The computer-implemented method of claim 2 , wherein the first response is output visually, the method further comprising:
determining whether a display period has elapsed since the first response was output visually; and
clearing the visual output of the first response after the elapse of the display period.
5. The computer-implemented method of claim 1 , further comprising:
responsive to identifying an absence of serial constraints between an earlier-received query and a later-received query, outputting a response to the later-received query before a response to the earlier-received query.
6. The computer-implemented method of claim 1 , wherein executing the first query involves performing an action.
7. A non-transitory computer-readable storage medium storing instructions that when executed by a computer processor perform actions comprising:
obtaining an audio input stream;
detecting in the audio input stream a beginning of a first utterance;
detecting in the audio input stream an end of the first utterance;
responsive to detecting the end of the first utterance:
starting to process the first utterance to recognize a first query;
starting to execute the first query to determine a first response;
continuing to process the audio input stream; and
detecting in the audio input stream a beginning of a second utterance;
responsive to detecting the end of the second utterance:
starting to process the second utterance to recognize a second query;
determining that there is a serial constraint between the first query and the second query; and
responsive to determining that there is a serial constraint between the first query and the second query, disallowing execution of the second query before execution of the first query.
8. The non-transitory computer-readable storage medium of claim 7 , wherein the first response is output visually, the actions further comprising:
determining whether a display period has elapsed since the first response was output visually; and
clearing the visual output of the first response after the elapse of the display period.
9. The non-transitory computer-readable storage medium of claim 7 , wherein executing the first query involves performing an action.
10. A computer system comprising:
at least a computer processor; and
a computer-readable storage medium storing instructions that when executed by at least the computer processor perform actions comprising:
obtaining an audio input stream;
detecting in the audio input stream a beginning of a first utterance;
detecting in the audio input stream an end of the first utterance;
responsive to detecting the end of the first utterance:
starting to process the first utterance to recognize a first query;
starting to execute the first query to determine a first response;
continuing to process the audio input stream; and
detecting in the audio input stream a beginning of a second utterance; and
responsive to detecting the end of the second utterance:
starting to process the second utterance to recognize a second query;
determining that there is not a serial constraint between the first query and the second query; and
responsive to determining that there is not a serial constraint between the first query and the second query, starting execution of the second query before the end of the execution of first query.
11. The computer system of claim 10 , the actions further comprising outputting the second response before the first response.
12. The computer system of claim 11 , wherein outputting the second response is performed concurrently with executing the first query.
13. The computer system of claim 11 , wherein the first response is output visually, the actions further comprising:
determining whether a display period has elapsed since the first response was output visually; and
clearing the visual output of the first response after the elapse of the display period.
14. The computer system of claim 10 , the actions further comprising:
responsive to identifying an absence of serial constraints between an earlier-received query and a later-received query, outputting a response to the later-received query before a response to the earlier-received query.
15. The computer system of claim 10 , wherein executing the first query involves performing an action.
16. A computer-implemented method of generating a response to a spoken input, the computer-implemented method comprising:
obtaining an audio input stream;
detecting in the audio input stream a beginning of a first utterance;
detecting in the audio input stream an end of the first utterance;
responsive to detecting the end of the first utterance:
starting to process the first utterance to recognize a first query;
starting to execute the first query to determine a first response;
continuing to process the audio input stream; and
detecting in the audio input stream a beginning of a second utterance; and
responsive to detecting the end of the second utterance:
starting to process the second utterance to recognize a second query;
determining that there is a serial constraint between the first query and the second query; and
responsive to determining that there is a serial constraint between the first query and the second query, disallowing execution of the second query before execution of the first query.
17. The computer-implemented method of claim 16 , wherein executing the first query involves performing an action.