WO2012135210A2

WO2012135210A2 - Location-based conversational understanding

Info

Publication number: WO2012135210A2
Application number: PCT/US2012/030730
Authority: WO
Inventors: Larry Paul Heck; Madhusudan Chinthakunta; David Mitby; Lisa Stifelman
Original assignee: Microsoft Corporation
Priority date: 2011-03-31
Filing date: 2012-03-27
Publication date: 2012-10-04
Also published as: CN102737104B; EP2691875A2; WO2012135218A2; CN106383866A; CN102750311B; WO2012135157A2; EP2691877A4; EP2691875A4; CN102750270B; CN102737099A; CN102737104A; JP6105552B2; CN102750270A; CN102750311A; EP2691876A4; WO2012135791A3; CN102737101A; KR20140025362A; JP2014515853A; EP2691876A2

Abstract

Location-based conversational understanding may be provided. Upon receiving a query from a user, an environmental context associated with the query may be generated. The query may be interpreted according to the environmental context. The interpreted query may be executed and at least one result associated with the query may be provided to the user.

Description

LOCATION-BASED CONVERSATIONAL UNDERSTANDING

BACKGROUND

[001] Location-based conversational understanding may provide a mechanism for leveraging environmental contexts to improve query execution and results. Conventional speech recognition programs do not have techniques for leveraging information (e.g., speech utterances, geographical data, acoustic environments of certain locations, typical queries made from a particular location) from one user to another to improve the quality and accuracy of new queries from new and/or existing users. In some situations, speech- to-text conversions must be made without the benefit of using similar, potentially related queries to aid in understanding.

[002] Speech-to-text conversion (i.e., speech recognition) may comprise converting a spoken phrase into a text phrase that may be processed by a computing system. Acoustic modeling and/or language modeling may be used in modern statistic- based speech recognition algorithms. Hidden Markov models (HMMs) are widely used in many conventional systems. HMMs may comprise statistical models that may output a sequence of symbols or quantities. HMMs may be used in speech recognition because a speech signal may be viewed as a piecewise stationary signal or a short-time stationary signal. In a short-time (e.g., 10 milliseconds), speech may be approximated as a stationary process. Speech may thus be thought of as a Markov model for many stochastic purposes.

SUMMARY

[003] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter. Nor is this Summary intended to be used to limit the claimed subject matter's scope.

[004] Location-based conversational understanding may be provided. Upon receiving a query from a user, an environmental context associated with the query may be generated. The query may be interpreted according to the environmental context. The interpreted query may be executed and at at least one result associated with the query may be provided to the user.

[005] Both the foregoing general description and the following detailed description provide examples and are explanatory only. Accordingly, the foregoing general description and the following detailed description should not be considered to be restrictive. Further, features or variations may be provided in addition to those set forth herein. For example, embodiments may be directed to various feature combinations and sub-combinations described in the detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

[006] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate various embodiments of the present invention. In the drawings:

[007] FIG. 1 is a block diagram of an operating environment;

[008] FIG. 2 is a flow chart of a method for providing location-based

conversational understanding; and

[009] FIG. 3 is a block diagram of a system including a computing device.

DETAILED DESCRIPTION

[010] The following detailed description refers to the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the following description to refer to the same or similar elements. While embodiments of the invention may be described, modifications, adaptations, and other implementations are possible. For example, substitutions, additions, or modifications may be made to the elements illustrated in the drawings, and the methods described herein may be modified by substituting, reordering, or adding stages to the disclosed methods. Accordingly, the following detailed description does not limit the invention. Instead, the proper scope of the invention is defined by the appended claims.

[011] Location-based conversational understanding may be provided. For example, a speech-to-text system may be provided that correlates information from multiple users in order to improve the accuracy of the conversion and the results of queries included in the converted statement. Consistent with embodiments of the invention, a personal assistant program may receive speech-based queries from user(s) at multiple locations. Each query may be analyzed for acoustic and/or environmental characteristics, and such characteristics may be stored and associated with the location from which the query was received. For example, a query received from a user at a subway station may detect the presence of acoustic echoes off of tile walls and/or the background

environmental sounds of crowds or the subway train. These characteristics may then be known to be filtered out in future queries from that location allowing greater accuracy in the conversion of those queries. Consistent with embodiments of the invention, the location may be defined, for example by a global positioning system (GPS) location of the user, an area code associated with the user, a ZIP code associated with the user, and/or a proximity of the user to a landmark (e.g., a train station, a stadium, a museum, an office building, etc.).

[012] Processing the query may comprise adapting the query according to an acoustic model. For example, the acoustic model may comprise a background sound known to be present at a particular location. Applying the acoustic model may allow the query to be converted more accurately by ignoring irrelevant sounds. The acoustic model may also allow for altering the display of any results associated with the query. For example, in a particularly loud environment, results may be displayed on a screen rather than via audio. The environmental context may further be associated with an

understanding model to aid in the speech-to-text conversion. For example, the understanding model may comprise a Hidden Markov Model (HMM). The environmental context may further be associated with a semantic model to aid in executing the query. For example, the semantic model may comprise an ontology.

[013] Further, the subject of the query may be used to improve the results for future queries. For example, if users at the subway station query "when is the next one?", the personal assistant program may determine, over the course of several queries, that the user wants to know when the next train will arrive. This may be accomplished by asking for clarification of the query from a first user, and storing the clarification for use in the future. For another example, if one user queries "when is the next one?" and another user queries "when is the next train?", the program may correlate these queries and make the assumption that both users are requesting the same information.

[014] FIG. 1 is a block diagram of an operating environment 100 for providing location-based conversational understanding. Operating environment 100 may comprise a spoken dialog system (SDS) 110 comprising a personal assistant program 112, a speech- to-text converter 1 14, and a context database 116. Personal assistant program 1 12 may receive queries over a network 120 from a first plurality of users 130(A)-(C) at a first location 140 and/or a second plurality of users 150(A)-(C) at a second location 160.

Context database 116 may be operative to store context data associated with queries received from users such as first plurality of users 130(A)-(C) and/or second plurality of users 150(A)-(C). The context data may comprise acoustic and/or environmental characteristics and query context information such as query subjects, time/date of the query, user details, and/or the location from which the query was made. Consistent with embodiments of the invention, network 120 may comprise, for example, a private data network (e.g., an intranet), a cellular data network, and/or a public network such as the Internet.

[015] The agent may be associated with a spoken dialog system (SDS). Such systems enable people to interact with computers with their voice. The primary component that drives the SDS may comprise a dialog manager: this component manages the dialog-based conversation with the user. The dialog manager may determine the intention of the user through a combination of multiple sources of input, such as speech recognition and natural language understanding component outputs, context from the prior dialog turns, user context, and/or results returned from a knowledge base (e.g., search engine). After determining the intention, the dialog manager may take an action, such as displaying the final results to the user and/or continuing in a dialog with the user to satisfy their intent. The spoken dialog system may comprise a plurality of conversational understanding models, such as acoustic models associated with locations and/or a spoken language understanding model for processing speech-based inputs.

[016] FIG. 2 is a flow chart setting forth the general stages involved in a method

200 consistent with an embodiment of the invention for providing location-based conversational understanding. Method 200 may be implemented using a computing device 300 as described in more detail below with respect to FIG. 3. Ways to implement the stages of method 200 will be described in greater detail below. Method 200 may begin at starting block 205 and proceed to stage 210 where computing device 300 may receive a speech-based query from a user at a location. For example, user 130(A) may send a query to SDS 110 via a device such as a cellular telephone.

[017] From stage 210, method 200 may advance to stage 215 where computing device 300 may determine whether an environmental context associated with the location exists in the memory storage. For example, SDS 110 may identify the location from which the query was received (e.g., first location 140) and determine whether an environmental context associated with that location is present in context database 1 16.

[018] If no context associated with the location exists, method 200 may advance to stage 220 where computing device 300 may identify at least one acoustic interference in the speech-based query. For example, SDS 110 may analyze the audio of the query and identify background noise such as that associated with a large crowd around user 130(A) and/or of a passing train.

[019] Method 200 may then advance to stage 225 where computing device 300 may identify at least one subject associated with the speech-based query. For example, if the query comprises "When is the next arrival?", SDS 110 may identify train schedules as the subject of the query when the user is at a train station.

[020] Method 200 may then advance to stage 230 where computing device 300 may create a new environmental context associated with the location for storing in the memory storage. For example, SDS 110 may store the identified acoustic interference and query subject in context database 1 16 as being associated with the user's location.

[021 ] If a context associated with the location does exist, method 200 may advance to stage 235 where computing device 300 may load the environmental context associated with the location. For example, SDS 1 10 may load an environmental context as described above from context database 116.

[022] After creating the context at stage 240 or loading the context at stage 235, method 200 may then advance to stage 240 where computing device 300 may convert the speech-based query to a text-based query according to the environmental context. For example, SDS 110 may convert the speech-based query to a text-based query by applying a filter for removing the at least one acoustic interference associated with the

environmental context.

[023] Method 200 may then advance to stage 245 where computing device 300 may execute the text-based query according to the environmental context. For example, SDS 1 10 may execute the query (e.g., "When is the next arrival?") within a search domain (e.g., train schedules) associated with the at least one subject associated with the environmental context.

[024] Method 200 may then advance to stage 250 where computing device 300 may provide at least one result of the executed text-based query to the user. For example, SDS 110 may transmit a result to a device associated with user 130(A) (e.g., a cellular phone) for display. Method 200 may then end at stage 255.

[025] An embodiment consistent with the invention may comprise a system for providing location-based conversational understanding. The system may comprise a memory storage and a processing unit coupled to the memory storage. The processing unit may be operative to receive a query from a user, generate an environmental context associated with the query, interpret the query according to the environmental context, execute the interpreted query, and provide at least one result of the query to the user. The query may comprise, for example, a spoken query that the processing unit may be operative to convert into computer-readable text. Consistent with embodiments of the invention, the speech-to-text conversion may utilize a Hidden Markov Model algorithm comprising statistical weightings for various likely words associated with an

understanding model and/or semantic concepts associated with a semantic model. The processing unit may be operative to increase a statistical weighting of at least one predicted word according to at least one previous query received from the location, for example, and store that statistical weighting as part of the environmental context.

[026] The environmental context may comprise an acoustic model associated with a location from which the query was received. The processing unit may be operative to adapt the query according to at least one background sound out of the speech-based query according to the acoustic model. For example, the background sound (e.g., a train whistle) may be known to be present in spoken queries received from a given location (e.g., a train station). The background sound may be detected, measured for pitch, amplitude, and other acoustic characteristics. The query may be adapted to ignore such sounds, and the sound may be computed and stored for application to future queries from that location. The processing unit may be further operative to receive a second speech- based query from a second user and adapt the query to same the background sound out according to the updated acoustic model. The processing unit may be further operative to aggregate the environmental contexts associated with a plurality of queries from a plurality of users and store the aggregated environmental contexts associated with the location.

[027] Another embodiment consistent with the invention may comprise a system for providing location-based conversational understanding. The system may comprise a memory storage and a processing unit coupled to the memory storage. The processing unit may be operative to receive a speech-based query from a user at a location, load an environmental context associated with the location, convert the speech-based query to text according to the environmental context, execute the converted query according to the environmental context, and provide at least one result associated with the executed query to the user. The environmental context may comprise, for example, a time of at least one previous query, a date of at least one previous query, a subject of at least one previous query, a semantic model comprising an ontology, an understanding model, and an acoustic model of the location. The processing unit may be operative to adapt the query according to a known acoustic interference associated with the location. The processing unit may be further operative to store a plurality of environmental contexts associated with a plurality of locations aggregated according to a plurality of queries received from a plurality of users. The processing unit may be further operative to receive a correction to the converted text from the user and update the environmental context according to the correction. The processing unit may be further operative to receive a second speech-based query from the user at a second location, load a second environmental context associated with the second location, convert the second speech-based query to text according to the second environmental context, execute the converted query according to the second

environmental context, and provide at least one second result associated with the executed query to the user.

[028] Yet another embodiment consistent with the invention may comprise a system for providing a context-aware environment. The system may comprise a memory storage and a processing unit coupled to the memory storage. The processing unit may be operative to receive a speech-based query from a user at a location and determine whether an environmental context associated with the location exists in the memory storage. In response to determining that the environmental context does not exist, the processing unit may be operative to identify at least one acoustic interference in the speech-based query, identify at least one subject associated with the speech-based query, and create a new environmental context associated with the location for storing in the memory storage. In response to determining that the environmental context does exist, the processing unit may be operative to load the environmental context. The processing unit may then be operative to convert the speech-based query to a text-based query according to the environmental context, wherein being operative to convert the speech-based query to a text-based query according to the environmental context comprises being operative to apply a filter for removing the at least one acoustic interference associated with the environmental context, execute the text-based query according to the environmental context, wherein being operative to execute the text-based query according to the environmental context comprises being operative to execute the query wherein the at least one acoustic interference is associated with an acoustic model and wherein the at least one identified subject is associated with a semantic model associated with the environmental context, and provide at least one result of the executed text-based query to the user.

[029] FIG. 3 is a block diagram of a system including computing device 300. Consistent with an embodiment of the invention, the aforementioned memory storage and processing unit may be implemented in a computing device, such as computing device 300 of FIG. 3. Any suitable combination of hardware, software, or firmware may be used to implement the memory storage and processing unit. For example, the memory storage and processing unit may be implemented with computing device 300 or any of other computing devices 318, in combination with computing device 300.The aforementioned system, device, and processors are examples and other systems, devices, and processors may comprise the aforementioned memory storage and processing unit, consistent with embodiments of the invention. Furthermore, computing device 300 may comprise an operating environment for system 100 as described above. System 100 may operate in other environments and is not limited to computing device 300.

[030] With reference to FIG. 3, a system consistent with an embodiment of the invention may include a computing device, such as computing device 300. In a basic configuration, computing device 300 may include at least one processing unit 302 and a system memory 304. Depending on the configuration and type of computing device, system memory 304 may comprise, but is not limited to, volatile (e.g., random access memory (RAM)), non- volatile (e.g., read-only memory (ROM)), flash memory, or any combination. System memory 304 may include operating system 305, one or more programming modules 306, and may include personal assistant program 1 12. Operating system 305, for example, may be suitable for controlling computing device 300's operation. Furthermore, embodiments of the invention may be practiced in conjunction with a graphics library, other operating systems, or any other application program and is not limited to any particular application or system. This basic configuration is illustrated in FIG. 3 by those components within a dashed line 308.

[031] Computing device 300 may have additional features or functionality. For example, computing device 300 may also include additional data storage devices (removable and/or non-removable) such as, for example, magnetic disks, optical disks, or tape. Such additional storage is illustrated in FIG. 3 by a removable storage 309 and a non-removable storage 310. Computer storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, or other data. System memory 304, removable storage 309, and non-removable storage 310 are all computer storage media examples (i.e., memory storage.) Computer storage media may include, but is not limited to, RAM, ROM, electrically erasable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store information and which can be accessed by computing device 300. Any such computer storage media may be part of device 300. Computing device 300 may also have input device(s) 312 such as a keyboard, a mouse, a pen, a sound input device, a touch input device, etc. Output device(s) 314 such as a display, speakers, a printer, etc. may also be included. The aforementioned devices are examples and others may be used.

[032] Computing device 300 may also contain a communication connection 316 that may allow device 300 to communicate with other computing devices 318, such as over a network in a distributed computing environment, for example, an intranet or the Internet. Communication connection 316 is one example of communication media.

Communication media may typically be embodied by computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and includes any information delivery media. The term "modulated data signal" may describe a signal that has one or more characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media may include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media. The term computer readable media as used herein may include both storage media and communication media.

[033] As stated above, a number of program modules and data files may be stored in system memory 304, including operating system 305. While executing on processing unit 302, programming modules 306 (e.g., personal assistant program 1 12) may perform processes including, for example, one or more of method 200's stages as described above. The aforementioned process is an example, and processing unit 302 may perform other processes. Other programming modules that may be used in accordance with

embodiments of the present invention may include electronic mail and contacts applications, word processing applications, spreadsheet applications, database

applications, slide presentation applications, drawing or computer-aided application programs, etc.

[034] Generally, consistent with embodiments of the invention, program modules may include routines, programs, components, data structures, and other types of structures that may perform particular tasks or that may implement particular abstract data types. Moreover, embodiments of the invention may be practiced with other computer system configurations, including hand-held devices, multiprocessor systems, microprocessor- based or programmable consumer electronics, minicomputers, mainframe computers, and the like. Embodiments of the invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[035] Furthermore, embodiments of the invention may be practiced in an electrical circuit comprising discrete electronic elements, packaged or integrated electronic chips containing logic gates, a circuit utilizing a microprocessor, or on a single chip containing electronic elements or microprocessors. Embodiments of the invention may also be practiced using other technologies capable of performing logical operations such as, for example, AND, OR, and NOT, including but not limited to mechanical, optical, fluidic, and quantum technologies. In addition, embodiments of the invention may be practiced within a general purpose computer or in any other circuits or systems.

[036] Embodiments of the invention, for example, may be implemented as a computer process (method), a computing system, or as an article of manufacture, such as a computer program product or computer readable media. The computer program product may be a computer storage media readable by a computer system and encoding a computer program of instructions for executing a computer process. The computer program product may also be a propagated signal on a carrier readable by a computing system and encoding a computer program of instructions for executing a computer process. Accordingly, the present invention may be embodied in hardware and/or in software (including firmware, resident software, micro-code, etc.). In other words, embodiments of the present invention may take the form of a computer program product on a computer-usable or computer-readable storage medium having computer-usable or computer-readable program code embodied in the medium for use by or in connection with an instruction execution system. A computer-usable or computer-readable medium may be any medium that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[037] The computer-usable or computer-readable medium may be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or propagation medium. More specific computer-readable medium examples (a non-exhaustive list), the computer-readable medium may include the following: an electrical connection having one or more wires, a portable computer diskette, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CD-ROM). Note that the computer-usable or computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[038] Embodiments of the present invention, for example, are described above with reference to block diagrams and/or operational illustrations of methods, systems, and computer program products according to embodiments of the invention. The

functions/acts noted in the blocks may occur out of the order as shown in any flowchart. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved.

[039] While certain embodiments of the invention have been described, other embodiments may exist. Furthermore, although embodiments of the present invention have been described as being associated with data stored in memory and other storage mediums, data can also be stored on or read from other types of computer-readable media, such as secondary storage devices, like hard disks, floppy disks, or a CD-ROM, a carrier wave from the Internet, or other forms of RAM or ROM. Further, the disclosed methods' stages may be modified in any manner, including by reordering stages and/or inserting or deleting stages, without departing from the invention.

[040] All rights including copyrights in the code included herein are vested in and the property of the Applicant. The Applicant retains and reserves all rights in the code included herein, and grants permission to reproduce the material only in connection with reproduction of the granted patent and for no other purpose.

[041] While the specification includes examples, the invention's scope is indicated by the following claims. Furthermore, while the specification has been described in language specific to structural features and/or methodological acts, the claims are not limited to the features or acts described above. Rather, the specific features and acts described above are disclosed as example for embodiments of the invention.

Claims

CLAIMS WHAT IS CLAIMED IS:

1. A method for providing location-based conversational understanding, the method comprising:

receiving a query from a user;

generating an environmental context associated with the query;

interpreting the query according to the environmental context;

executing the interpreted query; and

providing at least one result of the query to the user.

2. The method of claim I, wherein interpreting the query comprises adapting at least one conversational understanding model according to at least one previous query received from the location and wherein the at least one conversational understanding model comprises an acoustic model associated with a location from which the query was received.

3. The method of claim 2, wherein interpreting the query further comprises adapting the query according to at least one background sound associated with the query according to the acoustic model and wherein the background sound is measured within the query.

4. The method of claim 3, further comprising storing the background sound as the acoustic characteristic associated with the location.

5. The method of claim I, further comprising:

aggregating the environmental contexts associated with a plurality of queries from a plurality of user;

storing the aggregated environmental contexts associated with the location;

receiving a second speech-based query from a second user; and

interpreting the second speech-based query according to the aggregated environmental contexts.

6. The method of claim 5, wherein interpreting the query comprises increasing a statistical weighting of at least one predicted semantic concept according to the aggregated environmental contextsassociated received from the location.

7. A computer-readable medium which stores a set of instructions which when executed performs a method for providing location-based conversational understanding, the method executed by the set of instructions comprising:

receiving a speech-based query from a user at a location;

loading an environmental context associated with the location;

converting the speech-based query to text according to the environmental context; executing the converted query according to the environmental context; and providing at least one result associated with the executed query to the user.

8. The computer-readable medium of claim 7, further comprising:

receiving a correction to the converted text from the user,; and

updating the environmental context according to the correction, wherein updating the environmental context comprises modifying a probability associated with at least one word of the correction for application to a future query.

9. The computer-readable medium of claim 7, further comprising:

receiving a second speech-based query from the user, at a second location;

loading a second environmental context associated with the second location;

converting the second speech-based query to text according to the second environmental context;

executing the converted query according to the second environmental context; and providing at least one second result associated with the executed query to the user.

10. A system for providing location-based conversational understanding, the system comprising:

a memory storage; and

a processing unit coupled to the memory storage, wherein the processing unit is operative to:

receive a speech-based query from a user at a location, determine whether an environmental context associated with the location exists in the memory storage,

in response to determining that the environmental context does not exist:

identify at least one acoustic interference in the speech-based query;

identify at least one subject associated with the speech-based query; and create a new environmental context associated with the location for storing in the memory storage, wherein the at least one acoustic interference is associated with an acoustic model and whereint the at least one identified subject is associated with a semantic model,

in response to determining that the environmental context does exist, load the environmental context,

convert the speech-based query to a text-based query according to the environmental context, wherein being operative to convert the speech-based query to a text-based query according to the environmental context comprises being operative to adapt the query according at least one acoustic interference associated with the environmental context,

execute the text-based query according to the environmental context, wherein being operative to execute the text-based query according to the environmental context comprises being operative to execute the query according to the semantic model associated with the environmental context, and

provide at least one result of the executed text-based query to the user.