Automotive infotainment system with spatially-cognizant applications that interact with a speech interface
An automotive processing unit includes an infotainment system having a speech interface, an application suite comprising one or more spatially-cognizant applications, and an automotive assistant that is configured to execute one or more of the spatially-cognizant applications. The speech interface is configured to receive a navigation announcement from a navigator and a touring announcement from one of the spatially-cognizant applications and, in response, to cause a spoken announcement to be made audible in a vehicle's cabin through a loudspeaker. The spoken announcement comprising content from at least one of the touring announcement and the navigation announcement.
1 . A method for using an automotive assistant to use natural language to communicate information to an occupant of a vehicle, said method comprising:
receiving first and second outputs from respective first and second applications, said first and second outputs comprising first text and second text, respectively;
using a natural-language processor, generating hybrid text using said first text and said second text, wherein said hybrid text comprises information to be presented to said occupant; and
communicating said hybrid text to said user of said automotive assistant,
wherein communicating said hybrid text to said occupant comprises processing said hybrid text and wherein processing said hybrid text comprises using a text-to-speech converter to produce hybrid audio, wherein said method further comprises using a loudspeaker to communicate said hybrid audio to said occupant,
wherein said first application is a navigation application and said second application is a touring application,
wherein said first application produces said first text independently of said second application producing said second text, and
wherein generating said hybrid text using said first text and said second text comprises determining an insertion point within said first text and inserting said second text at said insertion point and wherein determining said insertion point comprises determining said grammatical structure of said first text.
2 . The method of claim 1 , wherein generating said hybrid text using said first text and said second text comprises incorporating into said hybrid text at least some text that is missing from both said first text and said second text and wherein said hybrid text represents a combination of a meaning of said first text and a meaning of said second text.
3 . The method of claim 1 , wherein generating said hybrid text using said first text and said second text comprises generating a summary of said first text and using said summary in said hybrid text.
4 . The method of claim 1 , wherein said first output comprises a first constraint, wherein said second output comprises a second constraint, wherein generating said hybrid text comprises generating said hybrid text so as to conform to said first constraint and to said second constraint, wherein said first constraint and said second constraint are selected from the group consisting of a visibility constraint and an interference constraint.
5 . An apparatus comprising
a navigator that provides location information, said location information comprising information that is indicative of a geographic location of said vehicle,
an automotive processing unit that is integrated into a vehicle having a cabin, and a loudspeaker that radiates sound into said cabin, and plural microphones disposed at seats in said vehicle, said automotive processing unit comprising said navigator,
a processor,
an application suite comprising one or more spatially-cognizant applications, wherein said spatially-cognizant applications provide output that depends at least in part on location of said vehicle,
an automotive assistant that is configured to execute one or more of said spatially-cognizant applications,
an infotainment system that comprises said application suite,
a tour-guide application, said tour-guide application being among said spatially-cognizant applications,
a speech interface that is configured:
to receive a navigation announcement from said navigator and a touring announcement from said tour-guide application,
to use said processor to combine content from said navigation announcement and content from said touring announcement to form a hybrid announcement,
and to cause said hybrid announcement to be made audible in said cabin through said loudspeaker, said hybrid announcement comprising content from said touring announcement and content from said navigation announcement, said processor being a natural-language processor that is a constituent of said speech interface.
6 . The apparatus of claim 5 , wherein first and second announcements are selected from said group consisting of said navigation announcement and said touring announcement and wherein said processor identifies, based on grammatical structure of said first announcement, an insertion point for inserting content from said second announcement into said first announcement to form said hybrid announcement.
7 . The apparatus of claim 5 , wherein first and second announcements are selected from the group consisting of said navigation announcement and said touring announcement, wherein said processor identifies a sentence in said second announcement, said sentence having an end, and wherein said natural-language processor is configured to insert content from said first announcement at said end to form said hybrid announcement.
8 . The apparatus of claim 5 , wherein first and second announcements are selected from said group consisting of said navigation announcement and said touring announcement and wherein said processor is configured to insert said second announcement between first and second sentences in said first announcement, thereby forming said hybrid announcement.
9 . The apparatus of claim 5 , wherein first and second announcements are selected from said group consisting of said navigation announcement and said touring announcement, wherein said hybrid announcement includes content from said first and second announcements, wherein said processor that identifies a sentence in said second announcement, selects, as an insertion point, a beginning of said sentence, and inserts content from said first announcement into said second announcement at said insertion point to thereby form said hybrid announcement.
10 . The apparatus of claim 5 , wherein said processor is configured to interleave content from said touring announcement and said navigation announcement to form said hybrid announcement, whereby said hybrid announcement includes content from both announcements.
11 . The apparatus of claim 5 , wherein said processor is configured to form said hybrid announcement by inserting a phrase between said navigation announcement and said touring announcement.
12 . The apparatus of claim 5 , wherein said speech interface causes insertion of a syntactic cue between said navigation announcement and said touring announcement to form said hybrid announcement.
13 . The apparatus of claim 5 , wherein said processor is configured to combine touring content from said touring announcement and navigation content from said navigation announcement to form said hybrid announcement and to deliver said touring content and said navigation content in said hybrid announcement using different voices.
14 . The apparatus of claim 5 , wherein said speech interface comprises a summarizer and wherein said summarizer is configured to receive said touring announcement and to generate a summarized version thereof, wherein said summarized version is interleaved into said navigation announcement to form said hybrid announcement.
15 . The apparatus of claim 5 , wherein said navigation announcement and said touring announcement comprise corresponding first and second sentences and wherein said hybrid announcement comprises a third sentence that incorporates content from said first and second sentences.
16 . An apparatus for causing a natural language hybrid announcement to be spoken within a vehicle having an automotive assistant that is configured to execute one or more applications, said applications being spatially-cognizant applications, each of which provides an announcement that is based at least in part on a location of said vehicle, said applications comprising first and second applications, said apparatus comprising:
a navigator that provides location information, said location information comprising information that is indicative of a geographic location of said vehicle,
an automotive processing unit that is integrated into a cabin of the vehicle, and a loudspeaker that radiates sound into said cabin, and plural microphones disposed at seats in said vehicle, said automotive processing unit comprising said navigator,
a processor,
an application suite comprising the spatially-cognizant applications,
an infotainment system,
a tour-guide application, said tour-guide application being among said spatially-cognizant applications, and
a speech interface that is configured for:
receiving first and second announcements from said first and second applications, respectively, said first announcement comprising text,
forming a hybrid announcement based on said first and second announcements, said hybrid announcement being a natural-language announcement,
causing said hybrid announcement to be output via the said loudspeaker into the said cabin of said vehicle at a time and location that is constrained by said first and second announcements.