Abstract: Word lattices that are generated by an automatic speech recognition system are used to generate a modified word lattice that is usable by a spoken language understanding module. In one embodiment, the spoken language understanding module determines a set of salient phrases by calculating an intersection of the modified word lattice, which is optionally preprocessed, and a finite state machine that includes a plurality of salient grammar fragments.

Abstract: A large amount of human labor is required to transcribe and annotate a training corpus that is needed to create and update models for automatic speech recognition (ASR) and spoken language understanding (SLU). Active learning enables a reduction in the amount of transcribed and annotated data required to train ASR and SLU models. In one aspect of the present invention, an active learning ASR process and active learning SLU process are coupled, thereby enabling further efficiencies to be gained relative to a process that maintains an isolation of data in both the ASR and SLU domains.

Abstract: An active labeling process is provided that aims to minimize the number of utterances to be checked again by automatically selecting the ones that are likely to be erroneous or inconsistent with the previously labeled examples. In one embodiment, the errors and inconsistencies are identified based on the confidences obtained from a previously trained classifier model. In a second embodiment, the errors and inconsistencies are identified based on an unsupervised learning process. In both embodiments, the active labeling process is not dependent upon the particular classifier model.

Abstract: Systems and methods for annotating speech data. The present invention reduces the time required to annotate speech data by selecting utterances for annotation that will be of greatest benefit. A selection module uses speech models, including speech recognition models and spoken language understanding models, to identify utterances that should be annotated based on criteria such as confidence scores generated by the models. These utterances are placed in an annotation list along with a type of annotation to be performed for the utterances and an order in which the annotation should proceed. The utterances in the annotation list can be annotated for speech recognition purposes, spoken language understanding purposes, labeling purposes, etc. The selection module can also select utterances for annotation based on previously annotated speech data and deficiencies in the various models.

Abstract: Systems and methods for annotating speech data. The present invention reduces the time required to annotate speech data by selecting utterances for annotation that will be of greatest benefit. A selection module uses speech models, including speech recognition models and spoken language understanding models, to identify utterances that should be annotated based on criteria such as confidence scores generated by the models. These utterances are placed in an annotation list along with a type of annotation to be performed for the utterances and an order in which the annotation should proceed. The utterances in the annotation list can be annotated for speech recognition purposes, spoken language understanding purposes, labeling purposes, etc. The selection module can also select utterances for annotation based on previously annotated speech data and deficiencies in the various models.

Abstract: A method and apparatus for speaker recognition is provided. One embodiment of a method for determining whether a given speech signal is produced by an alleged speaker, where a plurality of statistical models (including at least one support vector machine) have been produced for the alleged speaker based on a previous speech signal received from the alleged speaker, includes receiving the given speech signal, the speech signal representing an utterance made by a speaker claiming to be the alleged speaker, scoring the given speech signal using at least two modeling systems, where at least one of the modeling systems is a support vector machine, combining scores produced by the modeling systems, with equal weights, to produce a final score, and determining, in accordance with the final score, whether the speaker is likely the alleged speaker.

Abstract: A method of generating a natural language model for use in a spoken dialog system is disclosed. The method comprises using sample utterances and creating a number of hand crafted rules for each call-type defined in a labeling guide. A first NLU model is generated and tested using the hand crafted rules and sample utterances. A second NLU model is built using the sample utterances as new training data and using the hand crafted rules. The second NLU model is tested for performance using a first batch of labeled data. A series of NLU models are built by adding a previous batch of labeled data to training data and using a new batch of labeling data as test data to generate the series of NLU models with training data that increases constantly. If not all the labeling data is received, the method comprises repeating the step of building a series of NLU models until all labeling data is received.

Abstract: An active labeling process is provided that aims to minimize the number of utterances to be checked again by automatically selecting the ones that are likely to be erroneous or inconsistent with the previously labeled examples. In one embodiment, the errors and inconsistencies are identified based on the confidences obtained from a previously trained classifier model. In a second embodiment, the errors and inconsistencies are identified based on an unsupervised learning process. In both embodiments, the active labeling process is not dependent upon the particular classifier model.

Abstract: A large amount of human labor is required to transcribe and annotate a training corpus that is needed to create and update models for automatic speech recognition (ASR) and spoken language understanding (SLU). Active learning enables a reduction in the amount of transcribed and annotated data required to train ASR and SLU models. In one aspect of the present invention, an active learning ASR process and active learning SLU process are coupled, thereby enabling further efficiencies to be gained relative to a process that maintains an isolation of data in both the ASR and SLU domains.

Abstract: Disclosed is a system and method of training a spoken language understanding module. Such a module may be utilized in a spoken dialog system. The method of training a spoken language understanding module comprises training acoustic and language models using a small set of transcribed data ST, recognizing utterances in a set Su that are candidates for transcription using the acoustic and language models, computing confidence scores of the utterances, selecting k utterances that have the smallest confidence scores from Su and transcribing them into a new set Si, redefining St as the union of St and Si, redefining Su as Su minus Si, and returning to the step of training acoustic and language models if word accuracy has not converged.

Abstract: Open-domain question answering is the task of finding a concise answer to a natural language question using a large domain, such as the Internet. The use of a semantic role labeling approach to the extraction of the answers to an open domain factoid (Who/When/What/Where) natural language question that contains a predicate is described. Semantic role labeling identities predicates and semantic argument phrases in the natural language question and the candidate sentences. When searching for an answer to a natural language question, the missing argument in the question is matched using semantic parses of the candidate answers. Such a technique may improve the accuracy of a question answering system and may decrease the length of answers for enabling voice interface to a question answering system.

Abstract: A system and a method are provided. A textual transcript of a recorded voice communication is received. Text from a non-voice communication is received. Information about the textual transcript of the recorded voice communication and the text from the non-voice communication is provided such that a user can manage a group of communications, based at least in part on the textual transcript of the recorded voice communication and the text from the non-voice communication.

Abstract: An apparatus and a method are provided for building a spoken language understanding model. Labeled data may be obtained for a target application. A new classification model may be formed for use with the target application by using the labeled data for adaptation of an existing classification model. In some implementations, the existing classification model may be used to determine the most informative examples to label.

Abstract: Utterance data that includes at least a small amount of manually transcribed data is provided. Automatic speech recognition is performed on ones of the utterance data not having a corresponding manual transcription to produce automatically transcribed utterances. A model is trained using all of the manually transcribed data and the automatically transcribed utterances. A predetermined number of utterances not having a corresponding manual transcription are intelligently selected and manually transcribed. Ones of the automatically transcribed data as well as ones having a corresponding manual transcription are labeled. In another aspect of the invention, audio data is mined from at least one source, and a language model is trained for call classification from the mined audio data to produce a language model.

Abstract: The invention relates to a system and method for gathering data for use in a spoken dialog system. An aspect of the invention is generally referred to as an automated hidden human that performs data collection automatically at the beginning of a conversation with a user in a spoken dialog system. The method comprises presenting an initial prompt to a user, recognizing a received user utterance using an automatic speech recognition engine and classifying the recognized user utterance using a spoken language understanding module. If the recognized user utterance is not understood or classifiable to a predetermined acceptance threshold, then the method re-prompts the user. If the recognized user utterance is not classifiable to a predetermined rejection threshold, then the method transfers the user to a human as this may imply a task-specific utterance. The received and classified user utterance is then used for training the spoken dialog system.

Abstract: A machine-readable medium may include a group of reusable components for building a spoken dialog system. The reusable components may include a group of previously collected audible utterances. A machine-implemented method to build a library of reusable components for use in building a natural language spoken dialog system may include storing a dataset in a database. The dataset may include a group of reusable components for building a spoken dialog system. The reusable components may further include a group of previously collected audible utterances. A second method may include storing at least one set of data. Each one of the at least one set of data may include ones of the reusable components associated with audible data collected during a different collection phase.

Abstract: A spoken dialog system configured to use a combined confidence score. A first confidence score, indicating a confidence level in a speech recognition result of recognizing an utterance, is provided. A second confidence level, indicating a confidence level of mapping the speech recognition result to an intent, is provided. The first confidence score and the second confidence score are combined to form a combined confidence score. A determination is made, with respect to whether to accept the intent, based on the combined confidence score.

Abstract: An apparatus and a method are provided for using semantic role labeling for spoken language understanding. A received utterance semantically parsed by semantic role labeling. A predicate or at least one argument is extracted from the semantically parsed utterance. An intent is estimated based on the predicate or the at least one argument. In another aspect, a method is provided for training a spoken language dialog system that uses semantic role labeling. An expert is provided with a group of predicate/argument pairs. Ones of the predicate/argument pairs are selected as intents. Ones of the arguments are selected as named entities. Mappings from the arguments to frame slots are designed.