US20090070113A1 - System for handling frequently asked questions in a natural language dialog service - Google Patents

System for handling frequently asked questions in a natural language dialog service Download PDF

Info

Publication number
US20090070113A1
US20090070113A1 US12/266,835 US26683508A US2009070113A1 US 20090070113 A1 US20090070113 A1 US 20090070113A1 US 26683508 A US26683508 A US 26683508A US 2009070113 A1 US2009070113 A1 US 2009070113A1
Authority
US
United States
Prior art keywords
question
answer
user
faq
table
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US12/266,835
Inventor
Narendra K. Gupta
Mazin G. Rahim
Giuseppe Riccardi
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nuance Communications Inc
Original Assignee
AT&T Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US37496102P priority Critical
Priority to US10/326,692 priority patent/US7197460B1/en
Priority to US11/675,166 priority patent/US7451089B1/en
Application filed by AT&T Corp filed Critical AT&T Corp
Priority to US12/266,835 priority patent/US20090070113A1/en
Publication of US20090070113A1 publication Critical patent/US20090070113A1/en
Assigned to NUANCE COMMUNICATIONS, INC. reassignment NUANCE COMMUNICATIONS, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: AT&T INTELLECTUAL PROPERTY II, L.P.
Application status is Abandoned legal-status Critical

Links

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/22Procedures used during a speech recognition process, e.g. man-machine dialogue
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16Sound input; Sound output
    • G06F3/167Audio in a user interface, e.g. using voice commands for navigating, audio feedback

Abstract

A voice-enabled help desk service is disclosed. The service comprises an automatic speech recognition module for recognizing speech from a user, a spoken language understanding module for understanding the output from the automatic speech recognition module, a dialog management module for generating a response to speech from the user, a natural voices text-to-speech synthesis module for synthesizing speech to generate the response to the user, and a frequently asked questions module. The frequently asked questions module handles frequently asked questions from the user by changing voices and providing predetermined prompts to answer the frequently asked question.

Description

    PRIORITY DOCUMENTS
  • The present application is a continuation of U.S. patent application Ser. No. 11/675,166, filed Feb. 15, 2007, which is a continuation of U.S. patent application Ser. No. 10/326,692, filed Dec. 19, 2002, which claims priority to provisional application No. 60/374,961, filed Apr. 23, 2002, the contents of which are incorporated herein by reference.
  • RELATED APPLICATIONS
  • This case is related to commonly assigned U.S. patent application Ser. No. 10/235,266, filed on Dec. 19, 2002, the contents of which are incorporated herein by reference (attorney docket number 2002-0093A). This case is related to commonly assigned U.S. patent application Ser. No. 10/235,295, filed Dec. 19, 2002, the contents of which are incorporated herein by reference (attorney docket number 2002-0093). This case is related to commonly assigned U.S. patent application Ser. No. 10/326,691, filed Dec. 19, 2002, the contents of which are incorporated herein by reference (attorney docket number 2002-0051). The present application is related to commonly assigned U.S. patent application Ser. No. 10/160,461 filed May 31, 2002, Attorney Docket Number 2001-0320, the contents of which are incorporated herein by reference.
  • BACKGROUND OF THE INVENTION
  • 1. Field of the Invention
  • The present invention relates to dialog systems and more specifically to an extended spoken language understanding module for handling frequently asked questions.
  • 2. Discussion of Related Art
  • Voice-enabled applications are becoming more widespread as automatic speech recognition (ASR), spoken language understanding (SLU), dialog management (DM) and text-to-speech (TTS) synthesizers improve. These voice-enabled applications represent an evolution of traditional help desks that are currently available on the web or supported by human agents. The goals of a voice-enabled help desk include call routing to appropriate agents or departments, providing a wealth of information about various products and services, and conducting problem solving or trouble shooting.
  • Speech and language processing technologies have the potential of automating a variety of customer care services in large industry sectors such as telecommunications, insurance, finance, travel, etc. In an effort to reduce the cost structure of customer care services, many of these industries are depending more heavily on complex Interactive Voice Response (IVR) menus for either automating an entire transaction or for routing callers to an appropriate agent or department. Several studies have shown that the “unnatural” and poor user interfaces of such menus tend to confuse and frustrate callers, preventing the callers from accessing information, let alone obtaining, in many cases, obtaining the desired service they expect. For example, studies show that over 53% of surveyed consumers say that automated IVR systems are the most frustrating part of customer service. In one survey, 46% of consumers dropped their credit card provider and 30% of them dropped their phone company provider due to poor customer care.
  • The advent of speech and language technologies have the potential for improving customer care not only by cutting the huge cost of running call centers in general but also by providing a more natural communication mode for conversing with users without requiring them to navigate through a laborious touch-tone menu. This has the effect of improving customer satisfaction and increasing customer retention rate. These values, which collectively form the foundation for an excellent customer care experience, have been evident in the AT&T Call Routing “How May I Help You” service that provides national consumer services via an automated spoken dialog system.
  • Soon, speech and language technologies will play a more pivotal role in customer care service and in help desk applications where the objectives include call routing and accessing information, as well as solving technical problems, sales, recommendations, and trouble shooting. Many computing and telecommunication companies today provide some form of a help desk service through either the World Wide Web or using a human agent. There is an opportunity for spoken natural language interfaces to play a much bigger role in this industry.
  • FIG. 1 illustrates the basic components required for human-computer interactive spoken dialog systems 10. The customer 12 speaks and provides an audible voice request. An automatic speech recognition (ASR) module 14 recognizes the speech and provides the text of the speech to a spoken language understanding (SLU) module 16 that parses the natural language input into relevant information to determine the substance of the customer inquiry. A dialog manager (DM) 18 receives the information regarding what the customer asked and generates the substance of the response, which is transmitted to a language generator 20 for generating the text of the response. The response text is transmitted to a text-to-speech (TTS) module 22 for generating a synthetic voice that “speaks” the response to the customer 12.
  • Further, some systems that are deployed are programmed to follow a particular dialog flow to lead the customer to the proper destination or information. Often, various consumers will have common questions that are asked that perhaps may be outside the designed dialog flow. Previous systems fail to adequately and efficiently handle these kinds of frequently asked questions.
  • Current technologies fail to enable companies to afford generating automated help desks. Handcrafted systems require manual training, segmenting and labeling of data in preparation for the voice user interface in the particular domain of the company. The data required for handcrafted systems may comprise hours and hours of scripted dialog with humans and the computer. The scripted computer-human interactions are studied and processed in a labor-intensive manner to train the new spoken dialog service. Such systems are time-consuming and costly to build, thus effectively preventing many companies from participating and receiving the improved customer care service that can be provided.
  • SUMMARY OF THE INVENTION
  • What is needed in the art is a help desk service that provides a more natural information exchange between a user and the help desk that increases customer satisfaction. An advantage of the present invention is to enable a natural languages help desk application that provides increased ease of use for customers calling the help desk.
  • Another advantage of the present invention is to improve customer relationships with companies using a natural language help desk that understands and interacts with users in a more efficient and pleasant manner, especially when handling frequently asked questions. The help desk according to the present invention includes an extended spoken language understanding module that includes further features of handling frequently asked questions.
  • Embodiments of the present invention include systems, methods and computer-readable medium-stored instructions for providing a voice-enabled interface. In an exemplary embodiment, a voice-enabled help desk service comprises (1) an automatic speech recognition module having a general-purpose acoustic model and a domain-specific model, wherein the general-purpose model is used for bootstrapping at an initial deployment of the voice-enabled help desk service and the domain-specific model is used to adapt the automatic speech recognition module after deployment; (2) a spoken language understanding module that performs text normalization, entity extraction and semantic classification using a boosting approach that balances human-crafted rules with available data, the spoken language understanding module including a frequently asked questions module; (3) a dialog management module that comprises an interpreter, finite state machine engine and an action template; and (4) a natural voices text-to-speech synthesis module for synthesizing speech. The frequently asked questions module handles frequently asked questions from the user by changing voices and providing prompts and answers calculated to move the user from the question back to the main dialog.
  • Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The features and advantages of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. These and other features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth herein.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The foregoing advantages of the present invention will be apparent from the following detailed description of several embodiments of the invention with reference to the corresponding accompanying drawings, in which:
  • FIG. 1 illustrates the components of a general spoken dialog system;
  • FIG. 2 illustrates the general components used according to the present invention;
  • FIG. 3 illustrates an extended SLU module according to an aspect of the present invention;
  • FIG. 4 illustrates an exemplary dialog manager architecture for use in the present invention;
  • FIG. 5 illustrates a high-level user interface according to an aspect of the invention; and
  • FIG. 6 illustrates word accuracy results.
  • DETAILED DESCRIPTION OF THE INVENTION
  • The present invention may be understood according to exemplary embodiments disclosed herein. When creating a spoken dialog service, experimental results are presented in terms of recognition accuracy, understanding accuracy and call completion rate.
  • There are several technology requirements needed for a voice-enabled help desk application. FIG. 2 illustrates some of the core necessary components of such a system 30, including an automatic speech recognizer (ASR) 32 capable of recognizing large-vocabulary spontaneous speech, an extended language understanding module (SLU) 34 that parses the natural language input into relevant information, a dialog manager (DM) 36 that operates in a mixed-initiative mode, a language generation module 38 and a text-to-speech module (TTS) 40 capable of generating high-quality synthesized voices fonts. For example, AT&T Labs' Natural Voices speech technologies include customized natural voice TTS engines that may be used for a variety of applications.
  • The present disclosure provides improvements in various modules shown in FIG. 2 that improve the cost, deployment time, customer relationship capability, and overall user experience for help desk applications. The following description provides further information used for building help desk applications quickly and efficiently.
  • Those of skill in the art will appreciate that other embodiments of the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, and the like. Embodiments may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination thereof) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices. Accordingly, as used herein, the term “the system” will refer to any computer device or devices that are programmed to function and process the steps of the method.
  • When initially building a help desk application, an aspect of the process that takes the longest amount of time relates to transcription and annotation of domain-specific speech used in building the underlying recognition and understanding models. “Domain-specific speech” relates to the speech that is within the expected subject matter of a conversation with a person. For example, a travel-related web-site will have “domain-specific” speech such as “destination,” “one-way,” “round-trip” etc. When the help desk is deployed for a specific purpose or “domain,” the recognition and understanding modules can maintain a high level of accuracy.
  • When a help desk application is being developed for a company, information about that company must be gathered in order to train the recognition and understanding modules. The process for collecting and annotating speech data is not only expensive and laborious; it delays the deployment cycle of new services. The process according to the present invention of building help desk services begins by “mining” and “reusing” data and models. Data mining is done not only from other similar application domains such as telecommunications, insurance, airline, etc, but also from relevant emails, web pages and human/agent recordings. See U.S. patent application Ser. No. 10/326,691 (Attorney Docket 2002-0051) for more details regarding this process.
  • As part of the labeling process, sentences are annotated for speech understanding purposes. The sentences, as mentioned above, can come from any source such as emails or web-site data. This is done in two phases. The first phase includes identifying and marking domain-specific and domain-independent value entities such as phone numbers, credit card numbers, dates, times, service offerings, etc. The second phase includes associating each input sentence with one or more semantic tags (or classes) that identify the “meaning” of a user's request. These tags can be both general and application-specific and are structured in a hierarchical manner. For example, phrases such as “may I hear this again” and “yes what products do you offer” can be tagged as “discourse repea” and “discourse_yes, info_products” respectively.
  • Regarding the automatic speech recognition module 32, accurate recognition of spoken natural-language input for help desk applications requires two components: (1) a general-purpose subword-based acoustic model (or a set of specialized acoustic models combined together), and (2) a domain-specific stochastic language model (or a set of specialized language models). Creating help desk applications imposes two challenges in building these models. The present invention enables the ability to bootstrap the service during an initial deployment.
  • Once all the available data is obtained, and the phases of (1) identifying and marking domain-specific and domain-independent value entities and (2) associating each input sentence with one or more semantic tags (or classes) that identify the “meaning” of a user's request are complete, the system can be initially deployed. While the recognition accuracy is not always at an acceptable level at this stage, often enough data can be gathered from company emails and web-site data to at least deploy the help desk application.
  • The invention provides for adapting the help desk service as task-specific data becomes available through live use of the service. In the case of acoustic modeling, according to the present invention, the help desk ASR 32 engine initially uses a general-purpose context-dependent hidden Markov model. This model is then adapted using a Maximum a posteriori adaptation once the system is deployed in the field. See, e.g., Huang, Acero and Hon, Spoken Language Processing, Prentice Hall PTR (2001), pages 445-447 for more information regarding Maximum a posteriori adaptation.
  • When generating the ASR module 32, stochastic language models are preferred for providing the highest possibility of recognizing word sequences “said” by the user 42. The design of a stochastic language model is highly sensitive to the nature of the input language and the number of dialog contexts or prompts. A stochastic language module takes a probabilistic viewpoint of language modeling. See, e.g., Id., pages 554-560 for more information on stochastic language models. One of the major advantages of using stochastic language models is that they are trained from a sample distribution that mirrors the language patterns and usage in a domain-specific language. A disadvantage of such language models is the need for a large corpus of data when bootstrapping.
  • Task-specific language models tend to have biased statistics on content words or phrases and language style will vary according to the type of human-machine interaction (i.e., system-initiated vs. mixed initiative). While there are no universal statistics to search for, the invention seeks to converge to the task-dependent statistics. This is accomplished by using different sources of data to achieve fast bootstrapping of language models including language corpus drawn from, for example, domain-specific web site, language corpus drawn from emails (task-specific), and language corpus drawn a spoken dialog corpus (non-task-specific).
  • The first two sources of data (web sites and emails) can give a rough estimate of the topics related to the task. However the nature of the web and email data do not account for the spontaneous-speech speaking style. On the other hand, the third source of data can be a large collection of spoken dialog transcriptions from other dialog applications. In this case, although the corpus topics may not be relevant, the speaking style may be closer to the target help desk applications. The statistics of these different sources of data are combined via a mixture model paradigm to form an n-gram language model. See, e.g., Id., pages 558-560. These models are adapted once task-specific data becomes available.
  • Regarding the text-to-speech synthesis module 40, the extensive call flow in help desk applications to support information access and problem solving, and the need to rapidly create and maintain these applications, make it both difficult and costly to use live voice recordings for prompt generation. The TTS module 40 plays a critical role in the new breed of natural language services where up-to-the-minute information (e.g., time and weather) and customization to an individual's voice are necessary. According to the present invention, the TTS module 40 provides a large variety of distinctive voices and, within each voice, several speaking-styles of many different languages. This is helpful for “branding” of help desk services.
  • The TTS engine 40 uses AT&T Labs Natural Voices technology and voice fonts. See, e.g., M. Beutnagel and A. Conkie and J. Schroeter and Y. Stylanianou and A. Syrdal, “The AT&T Next Generation TTS System”, Joint Meeting of ASA, EAA and DAGA, 1999. Due to automation of the voice creation process, new and customized voice fonts can be created in less than a month. Including task-specific data (i.e., materials relevant to the application) can assure a higher quality TTS voice.
  • For example, the main voice font used in an exemplary help desk TTS engine 40, named “Crystal”, is trained with over 12 hours of interactive dialogs between human agents and customers. In the help desk application described below, over 8 different voice fonts have been used within the same application for presenting different languages and dialog contexts. Any number of different voice fonts may be provided in a help desk application.
  • Regarding the spoken language understanding (SLU) module 34, text normalization is an important step for minimizing “noise” variations among words and utterances. This has the potential of increasing the effective size of the training-set and improving the SLU 34 accuracy. The text normalization component is essentially based on using morphology, synonyms and other forms of syntactic normalization. The main steps include stemming, removal of disfluencies, non-alphanumeric and non-white space characters and using a synonyms dictionary.
  • An important functionality of an SLU module 34 is the ability to parse the input speech into meaningful phrases. Parsing for help desk applications is simplified to a process of identifying task-specific and task-independent entities (such as phone numbers, credit card number, product type, etc.). Each entity module is built using standard context-free grammar that can be represented by a finite state transducer. Following text normalization, the system identifies entities by composing each input text string with all active entity modules. For example, the sentence “my bill for January 2nd” is parsed as “my bill for <Date> January 2nd</Date>”. Entity extraction not only helps to provide the DM 36 with the necessary information to generate a desired action but also it provides some form of text normalization for improving the classification accuracy.
  • FIG. 3 illustrates an extended SLU module 34 for use in applications such as a help desk. The extended SLU 34 enables a more efficient means for responding to frequently asked questions. Customer care services that use natural language understanding technology to perform call routing are typically limited to 10-100 call types (categories). The low number of categories and the lack of deeper understanding of the input limit these systems from providing sufficiently detailed information about the task. Typically, it is therefore not possible to answer users' specific questions directly. Instead, most systems give very general answers, and users are left to deduce the answer to their specific questions.
  • In an aspect of the present invention, a method extends the capabilities of natural language dialog systems to directly answer users' specific questions without increasing the number of call types. Referring to FIG. 3, in a spoken dialog system, user utterances are transformed into text “Q” via an ASR 32. The SLU module 34 converts this text into a semantic representation, based on which a DM module 36 decides the next action to take or next prompt to play to the user. A classifier-based SLU 34 classifies the input using a classification module 43 into one or many pre-determined call types “C” that form the basis of the DM's 36 next action.
  • The extended SLU 34 processes the output from the classification module 43 by selecting data from a question and answer (QA) table 46 for the call type “C”. In one aspect of the invention, a plurality of QA tables is stored and available from which to select a particular QA table according to the call type “C.” Once the QA table is selected, the system searches 44 the QA table for a question similar to the current utterance or text “Q”. The QA table stores a series of corresponding answers to various questions. The SLU 34 selects the most similar question to the text “Q” from the QA table and outputs the corresponding answer 45. If a question with a high enough similarity is found in the QA table 46, the system selects the most similar question to “Q” from a list of questions in the table. The text “Q”, call type “C” and answer “A” are transmitted to the DM 36.
  • An example of the process of performing the FAQ similarity computation follows. In an initialization process, each question FAQj in the QA table is represented as a vector: (faqj1, faqj2, faqj3, - - - faqjm) assuming there are “m” unique words in all the questions in the QA table. The term faqji equals tfji times idfi, where tfji is the term frequency of wordi (1 if wordi is present, 0 if it is absent in FAQj). The term idfi is defined as an inverse document frequency of wordi and equals log N/(ni). The term N is the total number of questions in QA table and ni is the number of questions containing the wordi.
  • Sim ( Q , FAQ j ) = i q i · faq ji i ( q i ) 2 + c ( 0.5 ) 2 · i ( faq ij ) 2
  • For each incoming question Q to be answered, a system operating according to the present invention performs the following steps: (1) calculating a vector representation of the question Q (q1, q2, q3- - - qm), wherein qi equals tfi times idfi and tfi is 1 if wordi is present and 0 if it is absent in the question Q. The term idfi is the inverse document frequency of the wordi, calculated above during the initialization process; (2) finding the number of words “c” in the incoming question that are not found in the “m” unique words seen in the all FAQs in the QA table; (3) for each question FAQj in the QA table, calculating a similarity computation:
  • (4) selecting the FAQj that has maximum value of Sim(Q, FAQj); and (5) reading the corresponding answer from QA table. In this manner, a system operating according to the present invention will perform the similarity computation according to the above description of initialization and for each incoming question calculating and processing a similarity value for use in selecting the appropriate answer from the QA table. This particular similarity computation is an example of one of many similarity computational methods that may be used. Accordingly, the inventors recognize that unless specifically claimed, the particular similarity computation is not limited to the above formula.
  • Using the above-mentioned principles, a dialogue and classification of questions can take place. For example, an utterance like “Can your software run on Linux” is classified as a call type “tech_compatibility”. One could argue that it should be classified as Linux_compatibility, but such a classification would require Windows_compatibility, Unix_compatibility, etc. Classification base SLUs are ill suited for such an explosion in call types.
  • As another example of the process, suppose the phrase “Can I hear a female custom voice demonstration . . . hum . . . in Spanish?” is input to the SLU 34. The SLU 34 output may look like:
  • - <!-- Spanish Custom Female and default Demo (Male not available) -->
    - <spanishCustomFemale mode=“queued”>
    <prompt id=“intro” type=“TTS”
    src=“HelpDesk:Prompts:demo_1004_pc_v4”
    file=“../prompts/demo_1004_pc_v4.ul”>OK, I can show you our female voice
    in Spanish. This voice will be available with Release 1.1 on December
    first.</prompt>
    <prompt id=“voiceFemale” type=“TTS”
    src=“HelpDesk:Prompts:demo_1014_pc_v1”
    file=“../prompts/demo_1014_pc_v1.ul”>Here it is.</prompt>
    <audio src=“system:sounds:sil500ms” />
    <audio src=“HelpDesk:DemoVoices:spanish_fl” />
    <prompt id=“end” type=“TTS” src=“HelpDesk:Prompts:demo_1006_pc_v2”
    file=“../prompts/demo_1006_pc_v2.ul”>If you're interested in talking to
    someone about our Spanish voice, just ask for sales.</prompt>
    <var name=“rejection” value=“0” />
    <var name=“maxRejection” value=“1” />
    </spanishCustomFemale>

    The above XML output only serves as an example of one way the invention may be implemented. This particular implementation is not meant to be limiting of the scope of the invention.
  • In previous systems, the dialog manager responds to the question “Can your software run on Linux?” by playing a prompt that informs the user of all the platforms the software is compatible with, even though the user wanted to know only about Linux. An aspect of the present invention enables the system to generate a more appropriate answer to the question, such as: “Yes, our software can run on Linux”.
  • As shown in FIG. 3, the SLU 34 is extended to allow such responses by being provided with the QA table 46, possibly one for each call type. These can be viewed as frequently asked questions and their answers, possibly partitioned by call types.
  • To measure the similarity between user questions and questions in the QA Table 46, the system uses cosine similarity within the vector space model well known in the information retrieval field. According to the present invention, the system normalizes the vectors with the query length. As mentioned above, classifier-based SLUs and information retrieval technology are well known in the state of the art. This aspect of the invention combines classifier-based SLUs and information retrieval technology to answer direct questions.
  • The innovations disclosed herein allow development of effective spoken dialog-based help desks. They allow improvements in the question answering capability, simply by adding new questions and answers in the QA tables 46. Such capabilities can reduce the cost of customer support for companies and provide a mechanism for inserting immediate information into the system without involving any alteration to the ASR, SLU or the DM modules. Further information regarding the QA table 46 and answering frequently asked questions will be provided below with reference to FIG. 6.
  • An aspect of the present invention relates to semantic classification of utterances. The system categorizes each utterance into one or more semantic classes. A machine learning approach is taken for this task. The classifier 43 is trained using a corpus of collected utterances that have been annotated using a predefined set of semantic tags.
  • To train an exemplary classifier 43 according to the present invention, the system uses a technique called boosting. The basic idea of boosting is to combine many simple and moderately inaccurate prediction rules into a single rule that is highly accurate. Each of the base rules is trained on weighted versions of the original training set in which the “hardest” examples—i.e., those that are most often misclassified by the preceding rules—are given the greatest weight. The base rules are then combined into a single rule by taking a kind of majority vote. The first practical and still most widely studied boosting algorithm is Freund and Schapire's AdaBoost algorithm. See, e.g., Y. Friend and R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting”, Journal of Computer and Systems Sciences, 1997, for an overview of work on boosting.
  • In a preferred embodiment of the invention, the system uses an implementation of boosting developed by Schapire and Singer called BoosTexter. See, e.g., U.S. Patent Application No. 60/306,283, incorporated by reference above. In this implementation, each rule makes its predictions based simply on the presence or absence of a word or short phrase in the utterance. Like most machine-learning methods, boosting is heavily data driven, and so requires a good number of examples.
  • In developing help desk applications, it is often necessary to deploy the system before a sufficient number of examples have been collected. To get around this difficulty, the present invention uses human knowledge to compensate for the lack of data. In particular, the system uses a modification of boosting developed by Schapire et. al. that admits the direct incorporation of prior knowledge so that a classifier 43 is built by balancing human-crafted rules against what little data may be available. See the patent applications incorporated above for more information on using prior knowledge to boost the development of the classifier 43.
  • The human-built rules have a simple form and need not be perfectly accurate; for instance, one rule may state that if the word “demo” occurs in the utterance, then the user probably wants to hear a demonstration of some sort. Incorporating prior knowledge in a probabilistic fashion allows rapid deployment and a more effective way to add new tags throughout service evolution.
  • Regarding the DM module 36, significant challenges exist on how to build and easily maintain large-scale voice-enabled applications. The DM module 36 is designed according to the present invention to address the challenges of building voice-enabled applications. FIG. 4 illustrates an example of the architecture for the DM 36 of the present invention. The DM 36 is a particularly important issue for help desk applications where the nature of the information can be constantly changing. The complexity of the dialog modeling and the lack of adequate authoring tools can compromise the value and effectiveness of an automated help desk service.
  • The approach proposes, through general dialog patterns, a unified view to represent a human-machine dialog flow structure of commonly accepted reference models for mixed-initiative systems. A general engine operates on the semantic representation provided by the extended SLU 34 and current dialog context 51 (dialog strategy) to control the interaction flow. To describe the human-machine interaction, the system extends the traditional approach of finite state machines (FSM). FSMs are attractive mechanisms for dialog specification since they are (a) a direct translation of call flow specifications, (b) easy to augment with specific mixed-initiative interactions (c) practical to manage extensive dialog context. However, the use of FSM as mechanisms for managing the dialog flow is discussed, other means are also contemplated for this process. For example, dialog motivators may also be used as well as other mechanisms that are developed.
  • FIG. 4 illustrates the FSM engine 52 and FSM 64 in the DM 36 according to the present invention. The SLU module 34 processes the ASR output into, for example, the semantic representations in XML. The DM 36 shown here includes modules such as an interpreter 50, FSM engine 52, and action set templates 54. The FSM engine 52 controls the actions taken in response to the interpreter 50 output. Within the FSM structure, each information state provides support for general user interface patterns such as correction, start-over, repeat, confirmation, clarification, contextual help, and context shifts. Topic tracking is a feature that provides the infrastructure for rendering information. General conversation topics are managed by a subdialog that (a) handles, in a declarative way, new topics, (b) specifies the level of details per topic, and (c) allows context shift to take place at any point in the dialog.
  • According to the present invention, a dialog D is a tuple presented by the following expression D=<qo,Q,Θ,δ> where qoεQ is the initial state, Q is a finite set of nodes, FQ is a set of final nodes, Θ is the semantic representation produced by the interpreter 50, δ: Q×ΘΘQ is a transformation function q′=δ(q, i), where q, q′ are respectively the current and the next state and i is the current concept. The history mechanism can be modeled with the following tuple H=<T, C, φ> where T is the set of the current and past dialog turns including states variables and produced actions, c is the set of checkpoints or rollback states and φ: T×C→{true,false} is a Boolean function retuning true orfalse depending on the rollback strategy. Finally, the interpreter 50 maps to a tuple Θ=<
    Figure US20090070113A1-20090312-P00001
    ,H,Σ,σ>, where
    Figure US20090070113A1-20090312-P00001
    is a set of logical predicates, H is the dialog history, Σ is the set of input concepts and system events (e.g., semantic frames or slots), σ:
    Figure US20090070113A1-20090312-P00001
    ×H→Σ is a transformation function i=σ( r, h) where r is the set of rules that applies to the input frame and h is the current dialog history.
  • The interpreter 50 is responsible for providing a semantic interpretation of the concept categorization and the named entities provided by the SLU module 34. Logical predicates described in the rules file 62 allow the interpreter 50 to rank classes and assign a contextual interpretation to the input. An example of the output from the interpreter 50 follows:
  • if (slu.info_demo) {
    demo = getDemo(“female”,“custom”,“spanish”)
    n = history(demo)
    if (n>=0 && n< maxDemo) // demo already presented
    return demo[n]
    else return noMoreDemo
    }
  • The interpreter 50 also has access to the state variables 66, the dialog history 68 and user profiles 70. The user profile communicates with a user profile agent 56 that uses information 57 associated with the location of the user, the weather or local time for the user, or a black list of users or web content. The history mechanism model allows the DM 36 to capture situations where the request is under-specified or too general. For example, if the current topic has a missed mandatory attribute, the dialog engages a clarification sub-dialog in order to provide the missed information. Moreover, the history provides support for correction (rollback one dialog turn) or repeat requests. This last feature includes both explicit and semantic repeat requests. Situations where the user utters “please repeat that” or “what was the cost?” are then handled correctly by repeating the previous repeatable information.
  • An action template 54 receives actions as a result of the operation of the FSM engine 52. The action template 54 represents a template-based output generator. An XML-markup language describes the dialog actions (e.g., prompting, grammar activation, database queries, and variable values updates) and the topic structures. New topics and subtopics can be added, removed or updated at this level without changing the basic service logic. At run-time, the system translates the output by a XSL style sheet 72 either to Voice XML 60 (voice content through means such a telephone service) or to HTML 58 (text content) for output authoring. In this way, the presentation layer and the dialog structure for the topic sub dialog are completely separated from the service logic and are easy to maintain with traditional authoring tools.
  • An important aspect of the present invention relates to the User interface (“UI”). The UI is what the customer experiences when interacting with a system and plays an important role in providing quality service and user experience. There are two aspects in UI design for help desk applications: (a) usability goal—increasing the likelihood of call completion with minimal user confusion by supporting context shift in the dialog, providing information and help whenever necessary, and by learning how users interact with the system and propagating that knowledge into improving the various technology components; and (b) personality—creating an agent personality from synthesized speech that often lacks emotions. The system, according to an aspect of the present invention, uses a screenwriting dialog technique where a back story is created for the agents based on a set of desired personality traits (e.g., cheerful, trustworthy, calm, strong-willed, helpful, etc). As an example, a one-page description of the agent life history may be described and prompts are written “in-character”. A back story for an agent may be reflected in the agent making statements such as “I have to admit I'm having problems understanding you,” “Let's stop for just a quick second,” or “I honestly can't image using anything but the best!”
  • A plurality of different synthesized voices can be used conveying different information to the user. For example, an application may employ eight different voices for communicating different kinds of information to the user. The dialog strategy begins with the phrase “How May I Help You?” The system supports natural language input and context shift throughout the application. In the service application described below, several different opening prompts are deployed which help to analyze user feedback and how that is affected throughout the course of the dialog. For example, the top-level prompts can be of the category of demonstration and command-like hints: “Hi you are listening to AT&T Labs Natural Voices text to speech. I can show you a demo or answer your questions on topics like general information, pricing, or new languages. Don't hesitate to interrupt me at any time. Now, how can I help you?” Another category of top-level prompts is to promote a request for a demo only: “Hi you are listening to AT&T Labs Natural Voices text to speech. I can show you a demo or answer your questions. Now, how can I help you?” Or, the prompt can be designed to elicit questions regarding a demo or some hints on how to ask questions: “Hi you are listening to AT&T Labs Natural Voices text to speech. I can show you a demo or answer your questions. For example, you can ask me about the products we sell or how much they cost. Now, how can I help you?”
  • The various categories of top-level prompts and the examples above have been shown to mold and affect the user's requests. For example, studies show that the word content in user utterances varies for each category of top-level prompt. Accordingly, the particular phrases used in the dialog are chosen to prompt the user to use certain phrases or terms that are more likely to be understood by the system. FIG. 5 illustrates an exemplary high-level UI for a help desk application. In this case, the dialog begins with “How may I help you?” 110. The user may provide several responses. Suppose that the user says “I want to hear German.” A <play demo> routine 112 then runs to provide an example of German. Following this step, the system sends the user 42 to the prompt “anything else?” 114. If the user says “how much does it cost?”, then the system plays an <info prompt> and asks “Would you like to hear more?” 116. A “yes” returns the state to the <info_prompt> and question of wanting to know more and a “no” forwards the user to the prompt “anything else?” 114. From the prompt 114, a context shift 118 may occur where the user may ask for information or to hear another demonstration. Similarly, a context shift may occur between the prompt 116 and the state designated to play the demonstration 112.
  • An exemplary help desk application relates to a service for the AT&T Labs Natural Voices—a business that specializes in selling and marketing TTS products and voice fonts. The so-called TTS help desk took less than three months to design, develop and test. At one point following deployment, the TTS help desk received over 1000 calls per month from business customers. The service performs call routing to specialized agents (such as sales, technical, customer service) and provides information about the various products and services. The system also provides callers with a variety of demonstrations of the different voice fonts and languages.
  • An aspect of the present invention is the process of creating such help desk application. The initial data collection effort in building the help desk is primarily based on a large set of email interactions that took place prior to launching the business. Utterances stored from conversations associated with the products and services are manually extracted and annotated using a set of 62 broad semantic tags that describe the types and characteristics of the products and services the business was able to support. These tags are categorized into broader groupings such as agent, general information, help, technical and web site. Presented below are some benchmark results for the TTS help desk: (a) ASR, (b) question/answering, and (c) task completion rates, on a set of 1000 dialogs.
  • After the deployment of the TTS help desk, ASR results can be analyzed. Detailed analysis of the corpus shows that it exhibits a mixed sample of two language styles: key phrases and spontaneous language. The average number of user turns is 3.3 with 27% of users engaging longer interactions than average. Although there are roughly 75 possible prompts on each dialog turn, in studies the prompt contexts have been clustered into four types: generic, confirmation, language and help. Each context corresponded to a stochastic language model and was bootstrapped in three different ways: using web data, email data or an inventory of a human-machine (HD) database acquired from other dialog applications.
  • FIG. 6 shows 600 overall word accuracy of the TTS help desk system on the 1000 dialog interactions. These results show that the help desk was able to achieve 59% word accuracy without any formal data collection. When sufficient data was available (after 6 months from system deployment), the accuracy jumped to nearly 68%.
  • Help desk applications that are available on the web often provide an extensive list of Frequently Asked Questions (FAQs) to help users access detailed information in a straightforward manner. In text understanding, there are many systems that exist today that perform question/answering. For example, the AT&T 10-NAUT system (see www.ionaut.com) can provide answers to queries requesting entity information such as names and dates.
  • According to the present invention, the system incorporates a question/answering module to help users with task-specific FAQs. The FAQ module may be incorporated in the SLU 34 as shown in FIG. 4 or may be associated with another element of a spoken dialog system. In a preferred embodiment, the QA module is provided in the form of a QA table 46 as shown in FIG. 4, wherein the questions and answers are extracted from previous callers to the system. The accuracy of this module is improved by partitioning the table into smaller subsets, each corresponding to a semantic tag. During a test call, if a user asks a question which matches closely to one found in the QA table, the answer is automatically passed to the DM 36 along with any semantic tags (from the classifier) and entities. String matching is performed using cosine similarity within the vector space model well known in the information retrieval field. Better matching accuracy was observed if normalization of the vectors is carried out with the query length as opposed to the entire data set.
  • Regarding the question and answer results among some data collected by the inventors of the present invention, a small set of 250 questions from one specific tag were identified as potential FAQs and grouped into 81 distinct sets. Thus, for each answer there were potentially one or more questions. Given a question with a specific semantic tag, the task was to identify whether the system can identify the appropriate answer. The 81 sets of questions constituted as the training set were indexed using a vector space model. The test set consisted of 336 questions of which only 69 corresponded to valid questions, and the remaining are added to evaluate the robustness of the technique. At a given operating point, precision and recall were computed at 0.9 and 0.94, respectively, thus providing evidence of the effectiveness of the question-answering aspects of the present invention.
  • The benefits of the help desk system and method disclosed above have been shown since a bootstrapped system, after three months of deployment, illustrated an average 85% task completion rate. Although the functionalities of the system were continuously changing during that period of time, the successful statistic shows consistent improvement in the system. Results show that (a) the ASR accuracy which was initially at 59% through bootstrapping was improved to 68% following 6 months of system deployment; (b) question/answering results were at 0.9 and 0.94 for precision and recall, respectively; and (c) the latest release of the system performs at 84% and 85% semantic classification rate and task completion rate, respectively. These statistics only provide some information regarding the success of the approaches described above and are not meant to be limiting in any manner given that further research and refinement of the inventive concepts will improve the accuracy of help desk applications.
  • Embodiments within the scope of the present invention may also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or combination thereof) to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such connection is properly termed a computer-readable medium. Combinations of the above should also be included within the scope of the computer-readable media.
  • Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Computer-executable instructions also include program modules that are executed by computers in stand-alone or network environments. Generally, program modules include routines, programs, objects, components, and data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps.
  • Although the above description may contain specific details, they should not be construed as limiting the claims in any way. Other configurations of the described embodiments of the invention are part of the scope of this invention. Accordingly, the appended claims and their legal equivalents should only define the invention, rather than any specific examples given.

Claims (18)

1. A method of responding to user utterances in a spoken dialog service, the method comprising:
receiving a user utterance;
identifying a question and an answer selected from a question and answer table based on a similarity computation that calculates a similarity between the user utterance and at least one question in the question and answer table and a set of words in the user utterance that are not found in a set of unique words in the question and answer table; and
presenting an answer to the user utterance from the identified question and answer.
2. The method of claim 1, further comprising:
classifying the recognized text associated with the user utterance; and
selecting the question and answer table from a plurality of question and answer tables according at least to the classified text.
3. The method of claim 1, wherein the identified question and answer is output to a dialog management module and wherein the similarity computation further comprises:
Sim ( Q , FAQ j ) = i q i · faq ji i ( q i ) 2 + c ( 0.5 ) 2 · i ( faq ij ) 2
wherein the term faqji represents the questions in the question and answer table and the term qi represents the user utterance and c is the set of words.
4. The method of responding to utterances in a spoken dialog service of claim 1, wherein the identified question and answer output is selected according to a most similar question to the received utterance.
5. The method of responding to utterances of claim 1, wherein the answer is selected from the selected question and answer table as being associated with a question that is the most similar to the received utterance.
6. The method of responding to utterances of claim 1, wherein each question and answer table includes questions and associated answers.
7. A computer-readable medium storing a computer program having instructions for controlling a computing device to perform the steps of:
receiving a user utterance;
identifying a question and an answer selected from a question and answer table based on a similarity computation that calculates a similarity between the user utterance and at least one question in the question and answer table and a set of words in the user utterance that are not found in a set of unique words in the question and answer table; and
presenting an answer to the user utterance from the identified question and answer.
8. The computer-readable medium of claim 7, wherein the instructions further comprise:
classifying the recognized text associated with the user utterance; and
selecting the question and answer table from a plurality of question and answer tables according at least to the classified text.
9. The computer-readable medium of claim 7, wherein the instructions further comprise:
outputting the identified question and answer to a dialog management module and wherein the similarity computation further comprises:
Sim ( Q , FAQ j ) = i q i · faq ji i ( q i ) 2 + c ( 0.5 ) 2 · i ( faq ij ) 2
wherein the term faqji represents the questions in the question and answer table and the term qi represents the user utterance and c is the set of words.
10. The computer readable medium of claim 7, wherein the identified question and answer output is selected according to a most similar question to the received utterance.
11. The computer-readable medium of claim 7, wherein the answer is selected from the selected question and answer table as being associated with the question that is most similar to the received utterance.
12. The computer-readable medium of claim 7, wherein the question and answer table includes questions and associated answers.
13. A computing device for responding to utterances in a spoken dialog service, the computing device comprising:
a module configured to receive a user utterance;
a module configured to identify a question and answer selected from a question and answer table based on a similarity computation that calculates a similarity between the user utterance and at least one question in the question and answer table, a set of words in the user utterance that are not found in a set of unique words in the question and answer table; and
a module configured to present an answer to the user from the identified question and answer.
14. The computing device of claim 13, further comprising:
a module configured to classify recognized text associated with the user utterance; and
a module configured to select the question and answer table from a plurality of question and answer tables according at least to the classified text.
15. The computing device of claim 13, wherein the identified question and answer is output to a dialog management module, and wherein similarity computation further comprises:
Sim ( Q , FAQ j ) = i q i · faq ji i ( q i ) 2 + c ( 0.5 ) 2 · i ( faq ij ) 2
wherein the term faqji represents the questions in the question and answer table and the term qi represents the user utterance and c is the set of words.
16. The computing device of claim 13, wherein the identified question and answer output is selected according to a most similar question to the received utterance.
17. The computing device of claim 13, wherein the answer is selected from the selected question and answer table as being associated with a question that is the most similar to the received utterance.
18. The computing device of claim 13, wherein the question and answer table includes questions and associated answers.
US12/266,835 2002-04-23 2008-11-07 System for handling frequently asked questions in a natural language dialog service Abandoned US20090070113A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US37496102P true 2002-04-23 2002-04-23
US10/326,692 US7197460B1 (en) 2002-04-23 2002-12-19 System for handling frequently asked questions in a natural language dialog service
US11/675,166 US7451089B1 (en) 2002-04-23 2007-02-15 System and method of spoken language understanding in a spoken dialog service
US12/266,835 US20090070113A1 (en) 2002-04-23 2008-11-07 System for handling frequently asked questions in a natural language dialog service

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US12/266,835 US20090070113A1 (en) 2002-04-23 2008-11-07 System for handling frequently asked questions in a natural language dialog service

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US11/675,166 Continuation US7451089B1 (en) 2002-04-23 2007-02-15 System and method of spoken language understanding in a spoken dialog service

Publications (1)

Publication Number Publication Date
US20090070113A1 true US20090070113A1 (en) 2009-03-12

Family

ID=37886167

Family Applications (3)

Application Number Title Priority Date Filing Date
US10/326,692 Expired - Fee Related US7197460B1 (en) 2002-04-23 2002-12-19 System for handling frequently asked questions in a natural language dialog service
US11/675,166 Active US7451089B1 (en) 2002-04-23 2007-02-15 System and method of spoken language understanding in a spoken dialog service
US12/266,835 Abandoned US20090070113A1 (en) 2002-04-23 2008-11-07 System for handling frequently asked questions in a natural language dialog service

Family Applications Before (2)

Application Number Title Priority Date Filing Date
US10/326,692 Expired - Fee Related US7197460B1 (en) 2002-04-23 2002-12-19 System for handling frequently asked questions in a natural language dialog service
US11/675,166 Active US7451089B1 (en) 2002-04-23 2007-02-15 System and method of spoken language understanding in a spoken dialog service

Country Status (1)

Country Link
US (3) US7197460B1 (en)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090147932A1 (en) * 2007-09-19 2009-06-11 Genesis Financial Solutions, Inc. Enhanced programming model and controller for ivr
US20090187572A1 (en) * 2008-01-18 2009-07-23 Ives Edward M Systems and methods for providing access to instructional content
US20100217600A1 (en) * 2009-02-25 2010-08-26 Yuriy Lobzakov Electronic device and method of associating a voice font with a contact for text-to-speech conversion at the electronic device
US20110046951A1 (en) * 2009-08-21 2011-02-24 David Suendermann System and method for building optimal state-dependent statistical utterance classifiers in spoken dialog systems
WO2012094422A2 (en) * 2011-01-05 2012-07-12 Health Fidelity, Inc. A voice based system and method for data input
US8346563B1 (en) * 2012-04-10 2013-01-01 Artificial Solutions Ltd. System and methods for delivering advanced natural language interaction applications
US8379830B1 (en) * 2006-05-22 2013-02-19 Convergys Customer Management Delaware Llc System and method for automated customer service with contingent live interaction
US20130103404A1 (en) * 2011-10-21 2013-04-25 GM Global Technology Operations LLC Mobile voice platform architecture
US20130197907A1 (en) * 2011-10-18 2013-08-01 GM Global Technology Operations LLC Services identification and initiation for a speech-based interface to a mobile device
US20130197915A1 (en) * 2011-10-18 2013-08-01 GM Global Technology Operations LLC Speech-based user interface for a mobile device
US20140032206A1 (en) * 2012-07-30 2014-01-30 Microsoft Corpration Generating string predictions using contexts
US20140358631A1 (en) * 2013-06-03 2014-12-04 24/7 Customer, Inc. Method and apparatus for generating frequently asked questions
US20150024796A1 (en) * 2012-03-12 2015-01-22 Huawei Technologies Co., Ltd. Method for mobile terminal to process text, related device, and system
US20150057046A1 (en) * 2012-05-11 2015-02-26 Qualcomm Incorporated System and Methods for improving Page Decode Performance During Reading of System Information on a Multi-SIM Wireless Communication Device
US9229974B1 (en) 2012-06-01 2016-01-05 Google Inc. Classifying queries
US9326088B2 (en) 2011-10-21 2016-04-26 GM Global Technology Operations LLC Mobile voice platform architecture with remote service interfaces
US9336497B2 (en) 2014-01-06 2016-05-10 International Business Machines Corporation System and method for an expert question answer system from a dynamic corpus
US10235992B2 (en) * 2012-06-08 2019-03-19 Nvoq Incorporated Apparatus and methods using a pattern matching speech recognition engine to train a natural language speech recognition engine

Families Citing this family (230)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001013255A2 (en) * 1999-08-13 2001-02-22 Pixo, Inc. Displaying and traversing links in character array
US9076448B2 (en) 1999-11-12 2015-07-07 Nuance Communications, Inc. Distributed real time speech recognition system
US7050977B1 (en) 1999-11-12 2006-05-23 Phoenix Solutions, Inc. Speech-enabled server for internet website and method
US7392185B2 (en) * 1999-11-12 2008-06-24 Phoenix Solutions, Inc. Speech based learning/training system using semantic decoding
US7725307B2 (en) 1999-11-12 2010-05-25 Phoenix Solutions, Inc. Query engine for processing voice based queries including semantic decoding
US8645137B2 (en) 2000-03-16 2014-02-04 Apple Inc. Fast, language-independent method for user authentication by voice
ITFI20010199A1 (en) 2001-10-22 2003-04-22 Riccardo Vieri System and method for transforming text into voice communications and send them with an internet connection to any telephone set
US8126713B2 (en) * 2002-04-11 2012-02-28 Shengyang Huang Conversation control system and conversation control method
US7869998B1 (en) 2002-04-23 2011-01-11 At&T Intellectual Property Ii, L.P. Voice-enabled dialog system
US7398209B2 (en) * 2002-06-03 2008-07-08 Voicebox Technologies, Inc. Systems and methods for responding to natural language speech utterance
US7693720B2 (en) * 2002-07-15 2010-04-06 Voicebox Technologies, Inc. Mobile systems and methods for responding to natural language speech utterance
US8645122B1 (en) 2002-12-19 2014-02-04 At&T Intellectual Property Ii, L.P. Method of handling frequently asked questions in a natural language dialog service
US7412383B1 (en) * 2003-04-04 2008-08-12 At&T Corp Reducing time for annotating speech data to develop a dialog application
US7669134B1 (en) 2003-05-02 2010-02-23 Apple Inc. Method and apparatus for displaying information during an instant messaging session
JP2004343649A (en) * 2003-05-19 2004-12-02 Nec Corp Voice recognition and response system
US7454393B2 (en) * 2003-08-06 2008-11-18 Microsoft Corporation Cost-benefit approach to automatically composing answers to questions by extracting information from large unstructured corpora
US7386440B2 (en) * 2003-10-01 2008-06-10 International Business Machines Corporation Method, system, and apparatus for natural language mixed-initiative dialogue processing
US8799001B2 (en) * 2003-11-17 2014-08-05 Nuance Communications, Inc. Method and system for defining standard catch styles for speech application code generation
US7660400B2 (en) * 2003-12-19 2010-02-09 At&T Intellectual Property Ii, L.P. Method and apparatus for automatically building conversational systems
US7295981B1 (en) * 2004-01-09 2007-11-13 At&T Corp. Method for building a natural language understanding model for a spoken dialog system
US7421393B1 (en) 2004-03-01 2008-09-02 At&T Corp. System for developing a dialog manager using modular spoken-dialog components
US7412393B1 (en) * 2004-03-01 2008-08-12 At&T Corp. Method for developing a dialog manager using modular spoken-dialog components
US7430510B1 (en) * 2004-03-01 2008-09-30 At&T Corp. System and method of using modular spoken-dialog components
US7787601B2 (en) * 2004-03-10 2010-08-31 Convergys Customer Management Group, Inc. System and method for IVR signature detection and event management
JP4661074B2 (en) * 2004-04-07 2011-03-30 ソニー株式会社 The information processing system, information processing method, and a robot apparatus
US7633076B2 (en) 2005-09-30 2009-12-15 Apple Inc. Automated response to and sensing of user activity in portable devices
US7584103B2 (en) * 2004-08-20 2009-09-01 Multimodal Technologies, Inc. Automated extraction of semantic content and generation of a structured document from speech
US20130304453A9 (en) * 2004-08-20 2013-11-14 Juergen Fritsch Automated Extraction of Semantic Content and Generation of a Structured Document from Speech
US20060116880A1 (en) * 2004-09-03 2006-06-01 Thomas Gober Voice-driven user interface
JP3962767B2 (en) * 2004-10-08 2007-08-22 松下電器産業株式会社 Interactive support device
CA2525729A1 (en) * 2004-11-08 2006-05-08 At&T Corp. System and method for compiling rules created by machine learning program
US20060149553A1 (en) * 2005-01-05 2006-07-06 At&T Corp. System and method for using a library to interactively design natural language spoken dialog systems
US8185399B2 (en) 2005-01-05 2012-05-22 At&T Intellectual Property Ii, L.P. System and method of providing an automated data-collection in spoken dialog systems
US8478589B2 (en) 2005-01-05 2013-07-02 At&T Intellectual Property Ii, L.P. Library of existing spoken dialog data for use in generating new natural language spoken dialog systems
JP4924950B2 (en) * 2005-02-08 2012-04-25 日本電気株式会社 Question answering data editing device, question and answer data editing method, question answering data editing program
US7640160B2 (en) 2005-08-05 2009-12-29 Voicebox Technologies, Inc. Systems and methods for responding to natural language speech utterance
US7620549B2 (en) 2005-08-10 2009-11-17 Voicebox Technologies, Inc. System and method of supporting adaptive misrecognition in conversational speech
US7949529B2 (en) 2005-08-29 2011-05-24 Voicebox Technologies, Inc. Mobile systems and methods of supporting natural language human-machine interactions
US8677377B2 (en) 2005-09-08 2014-03-18 Apple Inc. Method and apparatus for building an intelligent automated assistant
JP4849662B2 (en) * 2005-10-21 2012-01-11 株式会社ピートゥピーエー Conversation control apparatus
JP4846336B2 (en) * 2005-10-21 2011-12-28 株式会社ピートゥピーエー Conversation control apparatus
JP4849663B2 (en) * 2005-10-21 2012-01-11 株式会社ピートゥピーエー Conversation control apparatus
DE102005061365A1 (en) * 2005-12-21 2007-06-28 Siemens Ag Background applications e.g. home banking system, controlling method for use over e.g. user interface, involves associating transactions and transaction parameters over universal dialog specification, and universally operating applications
US8301448B2 (en) 2006-03-29 2012-10-30 Nuance Communications, Inc. System and method for applying dynamic contextual grammars and language models to improve automatic speech recognition accuracy
US7831423B2 (en) * 2006-05-25 2010-11-09 Multimodal Technologies, Inc. Replacing text representing a concept with an alternate written form of the concept
US7505951B2 (en) * 2006-05-30 2009-03-17 Motorola, Inc. Hierarchical state machine generation for interaction management using goal specifications
US7797672B2 (en) * 2006-05-30 2010-09-14 Motorola, Inc. Statechart generation using frames
CA2652444A1 (en) * 2006-06-22 2007-12-27 Multimodal Technologies, Inc. Automatic decision support
US9620117B1 (en) 2006-06-27 2017-04-11 At&T Intellectual Property Ii, L.P. Learning from interactions for a spoken dialog system
US7860815B1 (en) * 2006-07-12 2010-12-28 Venkateswara Prasad Tangirala Computer knowledge representation format, system, methods, and applications
US7624075B2 (en) * 2006-09-15 2009-11-24 Microsoft Corporation Transformation of modular finite state transducers
US7627541B2 (en) * 2006-09-15 2009-12-01 Microsoft Corporation Transformation of modular finite state transducers
US8073681B2 (en) 2006-10-16 2011-12-06 Voicebox Technologies, Inc. System and method for a cooperative conversational voice user interface
US8918318B2 (en) * 2007-01-16 2014-12-23 Nec Corporation Extended recognition dictionary learning device and speech recognition system
CN101242452B (en) 2007-02-05 2013-01-23 国际商业机器公司 Method and system for automatic generation and provision of sound document
US7818176B2 (en) 2007-02-06 2010-10-19 Voicebox Technologies, Inc. System and method for selecting and presenting advertisements based on natural language processing of voice-based input
US7912828B2 (en) * 2007-02-23 2011-03-22 Apple Inc. Pattern searching methods and apparatuses
US8977255B2 (en) * 2007-04-03 2015-03-10 Apple Inc. Method and system for operating a multi-function portable electronic device using voice-activation
DE602008002017D1 (en) * 2007-05-22 2010-09-16 Honda Motor Co Ltd Device with speech comprehension
US8082151B2 (en) 2007-09-18 2011-12-20 At&T Intellectual Property I, Lp System and method of generating responses to text-based messages
US8074199B2 (en) * 2007-09-24 2011-12-06 Microsoft Corporation Unified messaging state machine
US9053089B2 (en) 2007-10-02 2015-06-09 Apple Inc. Part-of-speech tagging using latent analogy
US8165886B1 (en) 2007-10-04 2012-04-24 Great Northern Research LLC Speech interface system and method for control and interaction with applications on a computing system
WO2009055819A1 (en) * 2007-10-26 2009-04-30 Honda Motor Co., Ltd. Improving free-speech command classification for car navigation system
US8620662B2 (en) * 2007-11-20 2013-12-31 Apple Inc. Context-aware unit selection
US8140335B2 (en) 2007-12-11 2012-03-20 Voicebox Technologies, Inc. System and method for providing a natural language voice user interface in an integrated voice navigation services environment
US10002189B2 (en) 2007-12-20 2018-06-19 Apple Inc. Method and apparatus for searching using an active ontology
US9330720B2 (en) 2008-01-03 2016-05-03 Apple Inc. Methods and apparatus for altering audio output signals
US8327272B2 (en) 2008-01-06 2012-12-04 Apple Inc. Portable multifunction device, method, and graphical user interface for viewing and managing electronic calendars
US8868424B1 (en) * 2008-02-08 2014-10-21 West Corporation Interactive voice response data collection object framework, vertical benchmarking, and bootstrapping engine
WO2009102885A1 (en) * 2008-02-12 2009-08-20 Phone Through, Inc. Systems and methods for enabling interactivity among a plurality of devices
US8065143B2 (en) 2008-02-22 2011-11-22 Apple Inc. Providing text input using speech data and non-speech data
US7467090B1 (en) 2008-02-27 2008-12-16 International Business Machines Corporation Using web FAQ data for creating self-service speech applications
US8996376B2 (en) 2008-04-05 2015-03-31 Apple Inc. Intelligent text-to-speech conversion
US8732114B2 (en) * 2008-04-14 2014-05-20 Nuance Communications, Inc. Knowledge re-use for call routing
CA2665055C (en) * 2008-05-23 2018-03-06 Accenture Global Services Gmbh Treatment processing of a plurality of streaming voice signals for determination of responsive action thereto
CA2665014A1 (en) * 2008-05-23 2009-11-23 Accenture Global Services Gmbh Recognition processing of a plurality of streaming voice signals for determination of responsive action thereto
CA2665009C (en) * 2008-05-23 2018-11-27 Accenture Global Services Gmbh System for handling a plurality of streaming voice signals for determination of responsive action thereto
US9305548B2 (en) 2008-05-27 2016-04-05 Voicebox Technologies Corporation System and method for an integrated, multi-modal, multi-device natural language voice services environment
US8589161B2 (en) 2008-05-27 2013-11-19 Voicebox Technologies, Inc. System and method for an integrated, multi-modal, multi-device natural language voice services environment
US8464150B2 (en) 2008-06-07 2013-06-11 Apple Inc. Automatic language identification for dynamic text processing
US20100030549A1 (en) 2008-07-31 2010-02-04 Lee Michael M Mobile device having human language translation capability with positional feedback
US8768702B2 (en) 2008-09-05 2014-07-01 Apple Inc. Multi-tiered voice feedback in an electronic device
US8898568B2 (en) 2008-09-09 2014-11-25 Apple Inc. Audio user interface
US8676586B2 (en) * 2008-09-16 2014-03-18 Nice Systems Ltd Method and apparatus for interaction or discourse analytics
US20100066742A1 (en) * 2008-09-18 2010-03-18 Microsoft Corporation Stylized prosody for speech synthesis-based applications
US8712776B2 (en) 2008-09-29 2014-04-29 Apple Inc. Systems and methods for selective text to speech synthesis
US8352272B2 (en) * 2008-09-29 2013-01-08 Apple Inc. Systems and methods for text to speech synthesis
US8355919B2 (en) * 2008-09-29 2013-01-15 Apple Inc. Systems and methods for text normalization for text to speech synthesis
US8396714B2 (en) * 2008-09-29 2013-03-12 Apple Inc. Systems and methods for concatenation of words in text to speech synthesis
US8352268B2 (en) * 2008-09-29 2013-01-08 Apple Inc. Systems and methods for selective rate of speech and speech preferences for text to speech synthesis
US8583418B2 (en) 2008-09-29 2013-11-12 Apple Inc. Systems and methods of detecting language and natural language strings for text to speech synthesis
US8676904B2 (en) 2008-10-02 2014-03-18 Apple Inc. Electronic devices with voice command and contextual data processing capabilities
US8165887B2 (en) * 2008-12-08 2012-04-24 Nuance Communications, Inc. Data-driven voice user interface
US9959870B2 (en) 2008-12-11 2018-05-01 Apple Inc. Speech recognition involving a mobile device
US8990088B2 (en) * 2009-01-28 2015-03-24 Microsoft Corporation Tool and framework for creating consistent normalization maps and grammars
US8862252B2 (en) 2009-01-30 2014-10-14 Apple Inc. Audio user interface for displayless electronic device
US8401855B2 (en) * 2009-02-06 2013-03-19 Robert Bosch Gnbh System and method for generating data for complex statistical modeling for use in dialog systems
US8326637B2 (en) 2009-02-20 2012-12-04 Voicebox Technologies, Inc. System and method for processing multi-modal device interactions in a natural language voice services environment
US8380507B2 (en) 2009-03-09 2013-02-19 Apple Inc. Systems and methods for determining the language to use for speech generated by a text to speech engine
US8332225B2 (en) * 2009-06-04 2012-12-11 Microsoft Corporation Techniques to create a custom voice font
US10241752B2 (en) 2011-09-30 2019-03-26 Apple Inc. Interface for a virtual digital assistant
US20100312547A1 (en) * 2009-06-05 2010-12-09 Apple Inc. Contextual voice commands
US9858925B2 (en) 2009-06-05 2018-01-02 Apple Inc. Using context information to facilitate processing of commands in a virtual assistant
US9431006B2 (en) 2009-07-02 2016-08-30 Apple Inc. Methods and apparatuses for automatic speech recognition
US20110010179A1 (en) * 2009-07-13 2011-01-13 Naik Devang K Voice synthesis and processing
US8509396B2 (en) * 2009-09-24 2013-08-13 International Business Machines Corporation Automatic creation of complex conversational natural language call routing system for call centers
US9171541B2 (en) 2009-11-10 2015-10-27 Voicebox Technologies Corporation System and method for hybrid processing in a natural language voice services environment
WO2011059997A1 (en) * 2009-11-10 2011-05-19 Voicebox Technologies, Inc. System and method for providing a natural language content dedication service
US8682649B2 (en) 2009-11-12 2014-03-25 Apple Inc. Sentiment prediction from textual data
US20110110534A1 (en) * 2009-11-12 2011-05-12 Apple Inc. Adjustable voice output based on device status
US8600743B2 (en) 2010-01-06 2013-12-03 Apple Inc. Noise profile determination for voice-related feature
US8381107B2 (en) 2010-01-13 2013-02-19 Apple Inc. Adaptive audio feedback system and method
US8311838B2 (en) 2010-01-13 2012-11-13 Apple Inc. Devices and methods for identifying a prompt corresponding to a voice input in a sequence of prompts
US10276170B2 (en) 2010-01-18 2019-04-30 Apple Inc. Intelligent automated assistant
US20120022872A1 (en) * 2010-01-18 2012-01-26 Apple Inc. Automatically Adapting User Interfaces For Hands-Free Interaction
US9318108B2 (en) 2010-01-18 2016-04-19 Apple Inc. Intelligent automated assistant
US8682667B2 (en) 2010-02-25 2014-03-25 Apple Inc. User profiling for selecting user specific voice input processing information
US20110224982A1 (en) * 2010-03-12 2011-09-15 c/o Microsoft Corporation Automatic speech recognition based upon information retrieval methods
US8639516B2 (en) 2010-06-04 2014-01-28 Apple Inc. User-specific noise suppression for voice quality improvements
US8713021B2 (en) 2010-07-07 2014-04-29 Apple Inc. Unsupervised document clustering using latent semantic density analysis
US8719006B2 (en) 2010-08-27 2014-05-06 Apple Inc. Combined statistical and rule-based part-of-speech tagging for text-to-speech synthesis
US8719014B2 (en) 2010-09-27 2014-05-06 Apple Inc. Electronic device with text error correction based on voice recognition data
WO2012046562A1 (en) * 2010-10-06 2012-04-12 日本電気株式会社 Request acquisition support system in system development, request acquisition support method and recording medium
US8959102B2 (en) 2010-10-08 2015-02-17 Mmodal Ip Llc Structured searching of dynamic structured document corpuses
US8781836B2 (en) 2011-02-22 2014-07-15 Apple Inc. Hearing assistance system for providing consistent human speech
US9262612B2 (en) 2011-03-21 2016-02-16 Apple Inc. Device access using voice authentication
US9202465B2 (en) * 2011-03-25 2015-12-01 General Motors Llc Speech recognition dependent on text message content
US20120311584A1 (en) 2011-06-03 2012-12-06 Apple Inc. Performing actions associated with task items that represent tasks to perform
US10241644B2 (en) 2011-06-03 2019-03-26 Apple Inc. Actionable reminder entries
US10057736B2 (en) 2011-06-03 2018-08-21 Apple Inc. Active transport based notifications
US8812294B2 (en) 2011-06-21 2014-08-19 Apple Inc. Translating phrases from one language into another using an order-based set of declarative rules
US8706472B2 (en) 2011-08-11 2014-04-22 Apple Inc. Method for disambiguating multiple readings in language conversion
US8994660B2 (en) 2011-08-29 2015-03-31 Apple Inc. Text correction processing
US8762156B2 (en) 2011-09-28 2014-06-24 Apple Inc. Speech recognition repair using contextual information
US8958532B2 (en) * 2011-11-29 2015-02-17 International Business Machines Corporation Navigating an interactive voice response system
US8406384B1 (en) * 2012-01-18 2013-03-26 Nuance Communications, Inc. Universally tagged frequent call-routing user queries as a knowledge base for reuse across applications
US10134385B2 (en) 2012-03-02 2018-11-20 Apple Inc. Systems and methods for name pronunciation
US9483461B2 (en) 2012-03-06 2016-11-01 Apple Inc. Handling speech synthesis of content for multiple languages
US9280610B2 (en) 2012-05-14 2016-03-08 Apple Inc. Crowd sourcing information to fulfill user requests
US8775442B2 (en) 2012-05-15 2014-07-08 Apple Inc. Semantic search using a single-source semantic model
US9679568B1 (en) 2012-06-01 2017-06-13 Google Inc. Training a dialog system using user feedback
US9721563B2 (en) 2012-06-08 2017-08-01 Apple Inc. Name recognition system
WO2013185109A2 (en) 2012-06-08 2013-12-12 Apple Inc. Systems and methods for recognizing textual identifiers within a plurality of words
US8983840B2 (en) * 2012-06-19 2015-03-17 International Business Machines Corporation Intent discovery in audio or text-based conversation
US9495129B2 (en) 2012-06-29 2016-11-15 Apple Inc. Device, method, and user interface for voice-activated navigation and browsing of a document
US9576574B2 (en) 2012-09-10 2017-02-21 Apple Inc. Context-sensitive handling of interruptions by intelligent digital assistant
US9547647B2 (en) 2012-09-19 2017-01-17 Apple Inc. Voice-based media searching
US8935167B2 (en) 2012-09-25 2015-01-13 Apple Inc. Exemplar-based latent perceptual modeling for automatic speech recognition
US20140136210A1 (en) * 2012-11-14 2014-05-15 At&T Intellectual Property I, L.P. System and method for robust personalization of speech recognition
US8972243B1 (en) * 2012-11-20 2015-03-03 Amazon Technologies, Inc. Parse information encoding in a finite state transducer
US9607046B2 (en) 2012-12-14 2017-03-28 Microsoft Technology Licensing, Llc Probability-based state modification for query dialogues
US20140180695A1 (en) * 2012-12-25 2014-06-26 Microsoft Corporation Generation of conversation to achieve a goal
CN104969289A (en) 2013-02-07 2015-10-07 苹果公司 Voice trigger for a digital assistant
US9977779B2 (en) 2013-03-14 2018-05-22 Apple Inc. Automatic supplementation of word correction dictionaries
US9368114B2 (en) 2013-03-14 2016-06-14 Apple Inc. Context-sensitive handling of interruptions
US9733821B2 (en) 2013-03-14 2017-08-15 Apple Inc. Voice control to diagnose inadvertent activation of accessibility features
KR101904293B1 (en) 2013-03-15 2018-10-05 애플 인크. Context-sensitive handling of interruptions
WO2014144579A1 (en) 2013-03-15 2014-09-18 Apple Inc. System and method for updating an adaptive speech recognition model
AU2014233517B2 (en) 2013-03-15 2017-05-25 Apple Inc. Training an at least partial voice command system
US9607617B2 (en) * 2013-04-02 2017-03-28 Nuance Communications, Inc. Concept cloud in smart phone applications
WO2014197336A1 (en) 2013-06-07 2014-12-11 Apple Inc. System and method for detecting errors in interactions with a voice-based digital assistant
WO2014197334A2 (en) 2013-06-07 2014-12-11 Apple Inc. System and method for user-specified pronunciation of words for speech synthesis and recognition
US9582608B2 (en) 2013-06-07 2017-02-28 Apple Inc. Unified ranking with entropy-weighted information for phrase-based semantic auto-completion
WO2014197335A1 (en) 2013-06-08 2014-12-11 Apple Inc. Interpreting and acting upon commands that involve sharing information with remote devices
US10176167B2 (en) 2013-06-09 2019-01-08 Apple Inc. System and method for inferring user intent from speech inputs
KR101772152B1 (en) 2013-06-09 2017-08-28 애플 인크. Device, method, and graphical user interface for enabling conversation persistence across two or more instances of a digital assistant
JP2016521948A (en) 2013-06-13 2016-07-25 アップル インコーポレイテッド System and method for emergency call initiated by voice command
EP3047390A1 (en) * 2013-09-19 2016-07-27 Sysomos L.P. Systems and methods for actively composing content for use in continuous social communication
US9189742B2 (en) 2013-11-20 2015-11-17 Justin London Adaptive virtual intelligent agent
US10296160B2 (en) 2013-12-06 2019-05-21 Apple Inc. Method for extracting salient dialog usage from live data
US10181322B2 (en) * 2013-12-20 2019-01-15 Microsoft Technology Licensing, Llc Multi-user, multi-domain dialog system
US9620105B2 (en) 2014-05-15 2017-04-11 Apple Inc. Analyzing audio input for efficient speech and music recognition
US9502031B2 (en) 2014-05-27 2016-11-22 Apple Inc. Method for supporting dynamic grammars in WFST-based ASR
US9633004B2 (en) 2014-05-30 2017-04-25 Apple Inc. Better resolution when referencing to concepts
US9715875B2 (en) 2014-05-30 2017-07-25 Apple Inc. Reducing the need for manual start/end-pointing and trigger phrases
US9430463B2 (en) 2014-05-30 2016-08-30 Apple Inc. Exemplar-based natural language processing
US10078631B2 (en) 2014-05-30 2018-09-18 Apple Inc. Entropy-guided text prediction using combined word and character n-gram language models
US10170123B2 (en) 2014-05-30 2019-01-01 Apple Inc. Intelligent assistant for home automation
US9842101B2 (en) 2014-05-30 2017-12-12 Apple Inc. Predictive conversion of language input
WO2015184186A1 (en) 2014-05-30 2015-12-03 Apple Inc. Multi-command single utterance input method
US9785630B2 (en) 2014-05-30 2017-10-10 Apple Inc. Text prediction using combined word N-gram and unigram language models
US10289433B2 (en) 2014-05-30 2019-05-14 Apple Inc. Domain specific language for encoding assistant dialog
US9734193B2 (en) 2014-05-30 2017-08-15 Apple Inc. Determining domain salience ranking from ambiguous words in natural speech
US9760559B2 (en) 2014-05-30 2017-09-12 Apple Inc. Predictive text input
US9338493B2 (en) 2014-06-30 2016-05-10 Apple Inc. Intelligent automated assistant for TV user interactions
US9818400B2 (en) 2014-09-11 2017-11-14 Apple Inc. Method and apparatus for discovering trending terms in speech requests
US9626703B2 (en) 2014-09-16 2017-04-18 Voicebox Technologies Corporation Voice commerce
US9898459B2 (en) 2014-09-16 2018-02-20 Voicebox Technologies Corporation Integration of domain information into state transitions of a finite state transducer for natural language processing
US9886432B2 (en) 2014-09-30 2018-02-06 Apple Inc. Parsimonious handling of word inflection via categorical stem + suffix N-gram language models
US10074360B2 (en) 2014-09-30 2018-09-11 Apple Inc. Providing an indication of the suitability of speech recognition
US9646609B2 (en) 2014-09-30 2017-05-09 Apple Inc. Caching apparatus for serving phonetic pronunciations
US10127911B2 (en) 2014-09-30 2018-11-13 Apple Inc. Speaker identification and unsupervised speaker adaptation techniques
US9668121B2 (en) 2014-09-30 2017-05-30 Apple Inc. Social reminders
EP3207467A4 (en) 2014-10-15 2018-05-23 VoiceBox Technologies Corporation System and method for providing follow-up responses to prior natural language inputs of a user
US9711141B2 (en) 2014-12-09 2017-07-18 Apple Inc. Disambiguating heteronyms in speech synthesis
US9965514B2 (en) 2014-12-19 2018-05-08 Software Ag Usa, Inc. Techniques for real-time generation of temporal comparative and superlative analytics in natural language for real-time dynamic data analytics
US9865280B2 (en) 2015-03-06 2018-01-09 Apple Inc. Structured dictation using intelligent automated assistants
US9886953B2 (en) 2015-03-08 2018-02-06 Apple Inc. Virtual assistant activation
US9721566B2 (en) 2015-03-08 2017-08-01 Apple Inc. Competing devices responding to voice triggers
US9899019B2 (en) 2015-03-18 2018-02-20 Apple Inc. Systems and methods for structured stem and suffix language models
US9842105B2 (en) 2015-04-16 2017-12-12 Apple Inc. Parsimonious continuous-space phrase representations for natural language processing
US10083688B2 (en) 2015-05-27 2018-09-25 Apple Inc. Device voice control for selecting a displayed affordance
US10127220B2 (en) 2015-06-04 2018-11-13 Apple Inc. Language identification from short strings
US10101822B2 (en) 2015-06-05 2018-10-16 Apple Inc. Language input correction
US10186254B2 (en) 2015-06-07 2019-01-22 Apple Inc. Context-based endpoint detection
US10255907B2 (en) 2015-06-07 2019-04-09 Apple Inc. Automatic accent detection using acoustic models
US9977808B2 (en) * 2015-06-22 2018-05-22 Sap Se Intent based real-time analytical visualizations
US10108702B2 (en) 2015-08-24 2018-10-23 International Business Machines Corporation Topic shift detector
US9811519B2 (en) * 2015-09-24 2017-11-07 Conduent Business Services, Llc Generative discriminative approach for transactional dialog state tracking via collective matrix factorization
US9697820B2 (en) 2015-09-24 2017-07-04 Apple Inc. Unit-selection text-to-speech synthesis using concatenation-sensitive neural networks
AU2016346341A1 (en) * 2015-10-26 2018-05-10 24/7 Customer, Inc. Method and apparatus for facilitating customer intent prediction
US10049668B2 (en) 2015-12-02 2018-08-14 Apple Inc. Applying neural network language models to weighted finite state transducers for automatic speech recognition
US10223066B2 (en) 2015-12-23 2019-03-05 Apple Inc. Proactive assistance based on dialog communication between devices
JP2017167659A (en) 2016-03-14 2017-09-21 株式会社東芝 Machine translation device, method, and program
US10152507B2 (en) 2016-03-22 2018-12-11 International Business Machines Corporation Finding of a target document in a spoken language processing
US9934775B2 (en) 2016-05-26 2018-04-03 Apple Inc. Unit-selection text-to-speech synthesis based on predicted concatenation parameters
US9972304B2 (en) 2016-06-03 2018-05-15 Apple Inc. Privacy preserving distributed evaluation framework for embedded personalized systems
US10249300B2 (en) 2016-06-06 2019-04-02 Apple Inc. Intelligent list reading
US10049663B2 (en) 2016-06-08 2018-08-14 Apple, Inc. Intelligent automated assistant for media exploration
US10067938B2 (en) 2016-06-10 2018-09-04 Apple Inc. Multilingual word prediction
US10192552B2 (en) 2016-06-10 2019-01-29 Apple Inc. Digital assistant providing whispered speech
DK201670540A1 (en) 2016-06-11 2018-01-08 Apple Inc Application integration with a digital assistant
DK179415B1 (en) 2016-06-11 2018-06-14 Apple Inc Intelligent device arbitration and control
DK179343B1 (en) 2016-06-11 2018-05-14 Apple Inc Intelligent task discovery
WO2018023106A1 (en) 2016-07-29 2018-02-01 Erik SWART System and method of disambiguating natural language processing requests
US10043516B2 (en) 2016-09-23 2018-08-07 Apple Inc. Intelligent automated assistant
US10331791B2 (en) 2016-11-23 2019-06-25 Amazon Technologies, Inc. Service for developing dialog-driven applications
US10268680B2 (en) * 2016-12-30 2019-04-23 Google Llc Context-aware human-to-computer dialog
US10332518B2 (en) 2017-05-09 2019-06-25 Apple Inc. User interface for correcting recognition errors

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5895466A (en) * 1997-08-19 1999-04-20 At&T Corp Automated natural language understanding customer service system
US6377944B1 (en) * 1998-12-11 2002-04-23 Avaya Technology Corp. Web response unit including computer network based communication
US20030004838A1 (en) * 2001-06-29 2003-01-02 International Business Machines Corporation Information search system, information search method, call center system, server and information search apparatus
US6567805B1 (en) * 2000-05-15 2003-05-20 International Business Machines Corporation Interactive automated response system
US20040024739A1 (en) * 1999-06-15 2004-02-05 Kanisa Inc. System and method for implementing a knowledge management system
US6766316B2 (en) * 2001-01-18 2004-07-20 Science Applications International Corporation Method and system of ranking and clustering for document indexing and retrieval
US6922689B2 (en) * 1999-12-01 2005-07-26 Genesys Telecommunications Method and apparatus for auto-assisting agents in agent-hosted communications sessions
US7013308B1 (en) * 2000-11-28 2006-03-14 Semscript Ltd. Knowledge storage and retrieval system and method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5895466A (en) * 1997-08-19 1999-04-20 At&T Corp Automated natural language understanding customer service system
US6377944B1 (en) * 1998-12-11 2002-04-23 Avaya Technology Corp. Web response unit including computer network based communication
US20040024739A1 (en) * 1999-06-15 2004-02-05 Kanisa Inc. System and method for implementing a knowledge management system
US6922689B2 (en) * 1999-12-01 2005-07-26 Genesys Telecommunications Method and apparatus for auto-assisting agents in agent-hosted communications sessions
US6567805B1 (en) * 2000-05-15 2003-05-20 International Business Machines Corporation Interactive automated response system
US7013308B1 (en) * 2000-11-28 2006-03-14 Semscript Ltd. Knowledge storage and retrieval system and method
US6766316B2 (en) * 2001-01-18 2004-07-20 Science Applications International Corporation Method and system of ranking and clustering for document indexing and retrieval
US20030004838A1 (en) * 2001-06-29 2003-01-02 International Business Machines Corporation Information search system, information search method, call center system, server and information search apparatus

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9549065B1 (en) 2006-05-22 2017-01-17 Convergys Customer Management Delaware Llc System and method for automated customer service with contingent live interaction
US8379830B1 (en) * 2006-05-22 2013-02-19 Convergys Customer Management Delaware Llc System and method for automated customer service with contingent live interaction
US20090147932A1 (en) * 2007-09-19 2009-06-11 Genesis Financial Solutions, Inc. Enhanced programming model and controller for ivr
US20090187572A1 (en) * 2008-01-18 2009-07-23 Ives Edward M Systems and methods for providing access to instructional content
US20100217600A1 (en) * 2009-02-25 2010-08-26 Yuriy Lobzakov Electronic device and method of associating a voice font with a contact for text-to-speech conversion at the electronic device
US8645140B2 (en) * 2009-02-25 2014-02-04 Blackberry Limited Electronic device and method of associating a voice font with a contact for text-to-speech conversion at the electronic device
US8682669B2 (en) * 2009-08-21 2014-03-25 Synchronoss Technologies, Inc. System and method for building optimal state-dependent statistical utterance classifiers in spoken dialog systems
US20110046951A1 (en) * 2009-08-21 2011-02-24 David Suendermann System and method for building optimal state-dependent statistical utterance classifiers in spoken dialog systems
WO2012094422A2 (en) * 2011-01-05 2012-07-12 Health Fidelity, Inc. A voice based system and method for data input
WO2012094422A3 (en) * 2011-01-05 2014-04-10 Health Fidelity, Inc. A voice based system and method for data input
US9159322B2 (en) * 2011-10-18 2015-10-13 GM Global Technology Operations LLC Services identification and initiation for a speech-based interface to a mobile device
US20130197915A1 (en) * 2011-10-18 2013-08-01 GM Global Technology Operations LLC Speech-based user interface for a mobile device
US20130197907A1 (en) * 2011-10-18 2013-08-01 GM Global Technology Operations LLC Services identification and initiation for a speech-based interface to a mobile device
US9183835B2 (en) * 2011-10-18 2015-11-10 GM Global Technology Operations LLC Speech-based user interface for a mobile device
US20130103404A1 (en) * 2011-10-21 2013-04-25 GM Global Technology Operations LLC Mobile voice platform architecture
US9326088B2 (en) 2011-10-21 2016-04-26 GM Global Technology Operations LLC Mobile voice platform architecture with remote service interfaces
US9203877B2 (en) * 2012-03-12 2015-12-01 Huawei Device Co., Ltd. Method for mobile terminal to process text, related device, and system
US20150024796A1 (en) * 2012-03-12 2015-01-22 Huawei Technologies Co., Ltd. Method for mobile terminal to process text, related device, and system
US8346563B1 (en) * 2012-04-10 2013-01-01 Artificial Solutions Ltd. System and methods for delivering advanced natural language interaction applications
US20150057046A1 (en) * 2012-05-11 2015-02-26 Qualcomm Incorporated System and Methods for improving Page Decode Performance During Reading of System Information on a Multi-SIM Wireless Communication Device
US9294141B2 (en) * 2012-05-11 2016-03-22 Qualcomm Incorporated System and methods for improving page decode performance during reading of system information on a multi-SIM wireless communication device
US9229974B1 (en) 2012-06-01 2016-01-05 Google Inc. Classifying queries
US10235992B2 (en) * 2012-06-08 2019-03-19 Nvoq Incorporated Apparatus and methods using a pattern matching speech recognition engine to train a natural language speech recognition engine
US9195645B2 (en) * 2012-07-30 2015-11-24 Microsoft Technology Licensing, Llc Generating string predictions using contexts
US20140032206A1 (en) * 2012-07-30 2014-01-30 Microsoft Corpration Generating string predictions using contexts
WO2014197495A1 (en) * 2013-06-03 2014-12-11 24/7 Customer, Inc. Method and apparatus for generating frequently asked questions
US20140358631A1 (en) * 2013-06-03 2014-12-04 24/7 Customer, Inc. Method and apparatus for generating frequently asked questions
US9336497B2 (en) 2014-01-06 2016-05-10 International Business Machines Corporation System and method for an expert question answer system from a dynamic corpus
US9830557B2 (en) 2014-01-06 2017-11-28 International Business Machines Corporation System and method for an expert question answer system from a dynamic corpus

Also Published As

Publication number Publication date
US7197460B1 (en) 2007-03-27
US7451089B1 (en) 2008-11-11

Similar Documents

Publication Publication Date Title
Tur et al. Spoken language understanding: Systems for extracting semantic information from speech
Juang et al. Automatic recognition and understanding of spoken language-a first step toward natural human-machine communication
Gorin et al. How may I help you?
US5613036A (en) Dynamic categories for a speech recognition system
US6501833B2 (en) Method and apparatus for dynamic adaptation of a large vocabulary speech recognition system and for use of constraints from a database in a large vocabulary speech recognition system
US9240197B2 (en) Library of existing spoken dialog data for use in generating new natural language spoken dialog systems
Schnelle Context Aware Voice User Interfaces for Workflow Support
US7127393B2 (en) Dynamic semantic control of a speech recognition system
US6405170B1 (en) Method and system of reviewing the behavior of an interactive speech recognition application
US7177817B1 (en) Automatic generation of voice content for a voice response system
US8650026B2 (en) Methods for creating a phrase thesaurus
US5384892A (en) Dynamic language model for speech recognition
EP1163665B1 (en) System and method for bilateral communication between a user and a system
US6839667B2 (en) Method of speech recognition by presenting N-best word candidates
US8392188B1 (en) Method and system for building a phonotactic model for domain independent speech recognition
US8301436B2 (en) Semantic object synchronous understanding for highly interactive interface
US5390279A (en) Partitioning speech rules by context for speech recognition
US6363348B1 (en) User model-improvement-data-driven selection and update of user-oriented recognition model of a given type for word recognition at network server
US9082402B2 (en) Generic virtual personal assistant platform
US6937986B2 (en) Automatic dynamic speech recognition vocabulary based on external sources of information
US6832196B2 (en) Speech driven data selection in a voice-enabled program
US9684741B2 (en) Presenting search results according to query domains
US9002710B2 (en) System and method for applying dynamic contextual grammars and language models to improve automatic speech recognition accuracy
US6915246B2 (en) Employing speech recognition and capturing customer speech to improve customer service
US6754626B2 (en) Creating a hierarchical tree of language models for a dialog system based on prompt and dialog context

Legal Events

Date Code Title Description
STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION

AS Assignment

Owner name: NUANCE COMMUNICATIONS, INC., MASSACHUSETTS

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:AT&T INTELLECTUAL PROPERTY II, L.P.;REEL/FRAME:041512/0608

Effective date: 20161214