Service oriented speech recognition for in-vehicle automated interaction and in-vehicle user interfaces requiring minimal cognitive driver processing for same
View Patent ↗A system and method for implementing a server-based speech recognition system for multimodal automated interaction in a vehicle includes receiving, by a vehicle driver, audio prompts by an on-board human-to-machine interface and a response with speech to complete tasks such as creating and sending text messages, web browsing, navigation, etc. This service-oriented architecture is utilized to call upon specialized speech recognizers in an adaptive fashion. The human-to-machine interface enables completion of a text input task while driving a vehicle in a way that minimizes the frequency of the driver's visual and mechanical interactions with the interface, thereby eliminating unsafe distractions during driving conditions. After the initial prompting, the typing task is followed by a computerized verbalization of the text. Subsequent interface steps can be visual in nature, or involve only sound.
1. A method for implementing an interactive automated system, comprising:
using a telematics processing system located in proximity to a person:
receiving in a first interaction an indication of intent from the person;
thereafter, in a second interaction that is separate from the first interaction, receiving from the person a spoken utterance associated with the indicated intent, wherein the spoken utterance associated with the indicated intent matches intended text of a type of intent available to the person; and
processing the spoken utterance of the second interaction using the processing system;
transmitting the processed speech information to a remote data center using a wireless link;
analyzing the transmitted speech information;
based upon the indicated intent, selecting at least one optimal speech recognition engine from a set of speech recognition engines;
converting the analyzed speech information into packet data format to produce packet speech information;
using an internet-protocol transport network, transporting the packet speech information to the selected at least one optimal speech recognition engine and recognizing the converted speech information with the selected at least one optimal speech recognition engine;
retrieving recognition results and an associated confidence score from the selected at least one optimal speech recognition engine;
if the confidence score meets or exceeds a predetermined threshold for a best match, processing the recognition results to:
perform a search;
generate search results;
transport the search results to the processing system; and
present the search results to the person; and
if the confidence score is below the predetermined threshold, selecting at least one alternative optimal speech recognition engine to carry out recognition of the converted speech information.
2. The method according to claim 1 , wherein the processing system located in proximity to the person is a telematics processing system.
3. The method according to claim 1 , wherein the generated search results are in the form of a list of the search results.
4. The method according to claim 3 , wherein the list of search results is transported to the processing system and presented to the person.
5. The method according to claim 1 , wherein the at least one alternative optimal speech recognition engine is agent-assisted.
6. The method according to claim 1 , wherein the selected at least one optimal speech recognition engine is not local.
7. The method according to claim 1 , wherein the presentation of the search results is continued with the person prior to, or subsequent to, receiving the recognition results in an asynchronous manner.
8. The method according to claim 1 , wherein the presentation of the search results is continued with the person subsequent to receiving the recognition results in a synchronous manner.
9. The method according to claim 1 , further comprising logging packet data of the packet speech information and the recognition results for subsequent analysis.
10. The method according to claim 1 , wherein the processing system is located on-board a vehicle.
11. The method according to claim 10 , further comprising transporting vehicle location information along with the packet speech information to the selected at least one optimal speech recognition engine.
12. The method according to claim 11 , further comprising logging the vehicle location information for subsequent analysis.
13. The method according to claim 1 , wherein the indicated intent pertains to at least one of:
internet browsing; and
navigational information.
14. A method for implementing an interactive automated system, comprising:
using a telematics processing system located on-board a vehicle:
receiving in a first interaction an indication of intent from a driver of the vehicle;
thereafter, in a second interaction that is separate from the first interaction, receiving from the driver a spoken utterance associated with the indicated intent, wherein the spoken utterance associated with the indicated intent matches intended text of a type of intent available to the driver; and
processing the spoken utterance of the second interaction using the processing system;
transmitting the processed speech information to a remote data center using a wireless link;
analyzing the transmitted speech information;
based upon the indicated intent, selecting at least one optimal speech recognition engine from a set of speech recognition engines;
converting the analyzed speech information into packet data format to produce packet speech information;
using an internet-protocol transport network, transporting the packet speech information and vehicle location information to the selected at least one optimal speech recognition engine and recognizing the converted speech information with the selected at least one optimal speech recognition engine;
retrieving recognition results and an associated confidence score from the selected at least one optimal speech recognition engine;
if the confidence score meets or exceeds a predetermined threshold for a best match, processing the recognition results to:
perform a search;
generate search results;
transport the search results to the processing system; and
present the search results to the vehicle driver; and
if the confidence score is below the predetermined threshold, selecting at least one alternative optimal speech recognition engine to carry out recognition of the converted speech information.
15. The method according to claim 14 , wherein the at least one alternative optimal speech recognition engine is agent-assisted.
16. The method according to claim 14 , wherein the selected at least one optimal speech recognition engine is not local.
17. The method according to claim 14 , wherein the presentation of the search results is continued with the vehicle driver prior to, or subsequent to, receiving the recognition results in an asynchronous manner.
18. The method according to claim 14 , wherein the presentation of the search results is continued with the vehicle driver subsequent to receiving the recognition results in a synchronous manner.
19. The method according to claim 14 , wherein the indicated intent pertains to at least one of:
internet browsing; and
navigational information.
20. An interactive automated speech recognition system, comprising:
a telematics processing system located in proximity to a person;
a remote data center;
a wireless link that transmits processed speech information from the processing system to the remote data center,
wherein the processing system:
receives in a first interaction an indication of intent from the person;
thereafter, in a second interaction that is separate from the first interaction, receives from the person a spoken utterance associated with the indicated intent, wherein the spoken utterance associated with the indicated intent matches intended text of a type of intent available to the person; and
processes the spoken utterance of the second interaction and transmits the processed speech information to the remote data center using the wireless link, wherein the remote data center analyzes the transmitted processed speech information and converts the analyzed speech information into packet data format;
at least one optimal speech recognition engine selected to recognize the converted speech information, the at least one optimal speech recognition engine being selected from a set of speech recognition engines based upon the indicated intent;
an internet protocol transport network that transports the converted speech information to the selected at least one optimal speech recognition engine; and
wherein the selected at least one optimal speech recognition engine produces recognition results and an associated confidence score, whereby:
if the confidence score meets or exceeds a predetermined threshold for a best match, the recognition results are processed to:
perform a search;
generate search results;
transport the search results to the processing system; and
present the search results to the person; and
if the confidence score is below the predetermined threshold, at least one alternative optimal speech recognition engine is selected to carry out recognition of the converted speech information.