Abstract: An example method includes receiving, by a computational assistant executing at one or more processors, a representation of an utterance spoken at a computing device; identifying, based on the utterance, a task to be performed; determining a capability level of a first party (1P) agent to perform the task; determining capability levels of respective third party (3P) agents of a plurality of 3P agents to perform the task; responsive to determining that the capability level of the 1P agent does not satisfy a threshold capability level, that a capability level of a particular 3P agent of the plurality of 3P agents is a greatest of the determined capability levels, and that the capability level of the particular 3P agent satisfies the threshold capability level, selecting the particular 3P agent to perform the task; and performing one or more actions determined by the selected agent to perform the task.

Abstract: Implementations of the present specification disclose methods, apparatuses, and devices for generating word vectors. The method includes: obtaining individual words by segmenting a corpus; establishing a feature vector of each word based on n-ary characters; training a recurrent neural network based on the feature vectors of the obtained words and feature vectors of context words associated with the obtained words in the corpus; and generating a word vector for each obtained word based on the feature vector of the obtained word and the trained recurrent neural network.

Abstract: A convolutional neural network for classifying time series data uses a dynamic context selection. In one example a method includes receiving a plurality of inputs of different sizes at a convolutional neural network, applying convolution and pooling to each of the inputs to provide a plurality of outputs of different sizes, changing the size of each of the outputs to a selected uniform size, reshaping each of the outputs to a vector, and fully connecting the vectors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying data objects. One of the methods includes obtaining data that associates each term in a vocabulary of terms with a respective high-dimensional representation of the term; obtaining classification data for a data object, wherein the classification data includes a respective score for each of a plurality of categories, and wherein each of the categories is associated with a respective category label; computing an aggregate high-dimensional representation for the data object from high-dimensional representations for the category labels associated with the categories and the respective scores; identifying a first term in the vocabulary of terms having a high-dimensional representation that is closest to the aggregate high-dimensional representation; and selecting the first term as a category label for the data object.

Abstract: A method, system and computer program product for disambiguating meaning of terms. A natural language processing system obtains a first corpora of words used in a first sense and a second corpora of the same words that are used in a second sense. Each of these corpora of words may be associated with different linguistic domains. The natural language processing system generates a first and a second set of patterns using both the first and second corpora of words, respectively. A question passage is then received by the natural language processing system. The natural language processing system examines a word of interest in the question passage. The user is then notified that the word of interest is being used in the first sense or the second sense in response to identifying the word of interest in only the first set of patterns or the second set of patterns, respectively.

Abstract: Disclosed are an answer selection method for a question answering system. With the method and system, a sentence distributed representation of a question and an answer is generated by adopting a hierarchical length-adaptive neural network structure with a purpose of extracting high-quality sentence features by using different neural feature extractors according to a length of an input sentence. First, one word distributed representation is generated for each word in the input sentence by connecting one fixed word embedding and one finely-adjusted word embedding. Next, a BiLSTM-based feature extractor is used for a short sentence and a Transformer-based feature extractor is used for a long sentence. Finally, a sentence vector representation for measuring a correlation between a question and a candidate answer is generated by using an attention pooling layer considering an interaction between a question sentence and an answer sentence.

Abstract: An analysis device includes a memory and a processor configured to acquire a document for analyzing, generate a coded document including a plurality of coded words by coding the document for each word unit, determine a vector value according to a condition of appearance of a first coded phrase in the coded document, the first coded phrase corresponding to a combination of two or more coded words included in the plurality of coded words, and assign the vector value to the first coded phrase.

Abstract: A formula system can transform a formula natural language representation (“NLR”) into a representation which shows the formula in traditional mathematical notation. This transformation can include creating a state machine with transition mappings between states that match to initial parts of the NLR. These transition mappings can include global transition mappings that are first attempted to be matched to the beginning of the NLR and then state specific transition mappings can be matched to the NLR. The formula system can consume the NLR, transitioning from state to state as indicated by the transition mappings and removing the matched initial part from the NLR, until the NLR has been fully consumed. In some cases, the formula system can recursively or iteratively create additional state machines to consume portions of the NLR. Some states provide a result (e.g. portion of a formula representation) which are combined to create the final formula representation.

Abstract: Methods and apparatuses are described for analyzing unstructured computer text for identification and classification of complaint-specific interactions. A computer data stores unstructured text. A server computing device splits the unstructured text into phrases of words. The server generates a set of tokens from each phrase and removes tokens that are stopwords. The server generates a normalized sentiment score for each set of tokens. The server uses a rules-based classification engine to generate a rules-based complaint score for each set of tokens. The server uses an artificial intelligence machine learning model to generate a model-based complaint score for each set of tokens. The server determines determine whether each set of tokens corresponds to a complaint-specific interaction based upon the rules-based complaint score and the model-based complaint score.

Abstract: A speech recognition method and a speech recognition device are disclosed. The speech recognition method includes: obtaining features of a speech signal to be recognized; performing a path search in a search space generated by establishing a map according to the features to output a decoding result; judging whether a rejection is needed according to the decoding result; and when the rejection is needed, determining that a speech recognition result is the rejection, and when the rejection is not needed, obtaining the speech recognition result according to the decoding result. The method has a good recognition rejection effect.

Abstract: A computer-implemented method and supporting system transcribes spoken words being monitored from a telephonic interaction among two or more individuals. Telephonic interactions among the individuals are monitored, and at least two of the individuals are each assigned to a separate channel. While still being monitored, each of the channels is assigned a context-based speech recognition models, and in substantially real-time, the monitored telephonic interaction is transcribed from speech to text based on the different assigned models.

Abstract: The present disclosure generally relates to systems and processes for morpheme-based word prediction. An example method includes receiving a current word; determining a context of the current word based on the current word and a context of a previous word; determining, using a morpheme-based language model, a likelihood of a prefix based on the context of the current word; determining, using the morpheme-based language model, a likelihood of a stem based on the context of the current word; determining, using the morpheme-based language model, a likelihood of a suffix based on the context of the current word; determining a next word based on the likelihood of the prefix, the likelihood of the stem, and the likelihood of the suffix; and providing an output including the next word.

Abstract: Systems, methods, and devices for determining a target device for a voice command are provided. A voice command is detected at a plurality of devices. A weight is determined for the detected voice command at each device of the plurality of devices. The determined weight is exchanged among the plurality of devices. A highest weight among the exchanged weights is determined. The device associated with the highest weight is determined as the target device for the voice command.

Abstract: A voice processing system includes a voice acquisition unit that acquires an utterance of a user, a voice recognition unit that recognizes contents of the utterance acquired by the voice acquisition unit, an execution unit that executes processing based on a result of the voice recognition unit, and a determination unit that determines a candidate of a word to substitute a certain word. When the voice recognition unit recognizes a request including a first word, the execution unit executes the request including the first word and provides the user with a processing result. The determination unit determines a second word that is a substitute candidate of the first word. The execution unit also executes a request including the second word instead of the first word and stores a processing result in a storage unit.

Abstract: Disclosed is a method of extracting and using a semantic entity from a text message by an electronic device. The method includes: recognizing a text-based input; extracting a semantic entity from the text-based input; and providing the extracted semantic entity through an application in response to the application having a semantic setting that corresponds to the extracted semantic entity.

Abstract: A system, method, apparatus and computer readable medium for hierarchical speech recognition resolution. The method of hierarchical speech recognition resolution on a platform includes receiving a speech stream from a microphone. The speech stream is resolved using a lowest possible level automatic speech recognition (ASR) engine of multi-level ASR engines. The selection of the lowest possible level ASR engine is based on policies defined for the platform. If resolution of the speech stream is rated less than a predetermined confidence level, the resolution of the speech stream is pushed to a next higher-level ASR engine of the multi-level ASR engines until the resolution of the speech stream meets the predetermined confidence level without violating one or more policies.

Abstract: A mechanism is provided for improving predicate parses (or logical representations of a passage) using semantic knowledge. In response to encountering an ambiguous decision point during a syntactic analysis of a portion of natural language content, a candidate meaning of the ambiguous decision point is generated. Characteristics of the ambiguous decision point are evaluated based on a semantic knowledge base to determine a semantic meaning associated with the ambiguous decision point. A determination is made as to whether the semantic meaning supports or refutes the candidate meaning. In response to determining that the semantic meaning refutes the candidate meaning, the candidate meaning of the ambiguous decision point is overridden based on the semantic meaning to include the semantic meaning as a final meaning for the ambiguous decision point. The portion of natural language content is then processed based on the final meaning for the ambiguous decision point.

Abstract: A generation device includes a receiving unit and a generating unit. The receiving unit receives a model representing correspondence between one or more phonetic symbols and one or more words. The generating unit generates a first finite state transducer based on the model, the first finite state transducer at least including, as outgoing transitions from a first state representing transition destination of a first transition which has a first phonetic symbol of a predetermined type as input symbol, a second transition that has a second phonetic symbol, which is different than a particular symbol representing part or whole of input symbol of the first transition, as input symbol, and a third transition that has a third phonetic symbol, which represents the particular symbol or silence, as input symbol.

Abstract: A method, computer-readable medium, and system including a speech-to-text module to receive an input of speech including one or more words generated by a human and to output data including text, sentiment information, and other parameters corresponding to the speech input; a processing module like Artificial Intelligence to generate a reply to the speech input, the reply including a textual component, sentimental information associated with the textual component, and contextual information associated with the textual component; and a text-to-speech module to receive the textual component, sentimental information, and contextual information and to generate, based on the received textual component and its associated sentimental information and contextual information, a speech output including one or more spoken words, the spoken words to be presented with at least one of a pace, a tone, a volume, and an emphasis representative of the sentimental information and contextual information associated with the textual

Abstract: Aspects of the present invention disclose a method, computer program product, and system for parsing a domain-specific language (DSL) statement. The method includes one or more processors accessing a DSL statement that includes contracted phrases. The method further includes one or more processors identifying one or more contracted phrases in the DSL statement utilizing an annotated domain vocabulary for a DSL associated with the DSL statement and grammar rules for the DSL. The method further includes one or more processors determining expanded phrases corresponding to the identified one or more contracted phrases based on the annotated domain vocabulary and the grammar rules. The method further includes one or more processors creating an expanded abstract syntax tree (AST) that is representative of the DSL statement with the determined expanded phrases replacing the identified one or more contracted phrases.

Abstract: A method for controlling a drone includes receiving a request for information about a spatial location, generating data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: Disclosed herein are systems, methods, and computer-readable storage media for improving speech recognition accuracy using textual context. The method includes retrieving a recorded utterance, capturing text from a device display associated with the spoken dialog and viewed by one party to the recorded utterance, and identifying words in the captured text that are relevant to the recorded utterance. The method further includes adding the identified words to a dynamic language model, and recognizing the recorded utterance using the dynamic language model. The recorded utterance can be a spoken dialog. A time stamp can be assigned to each identified word. The method can include adding identified words to and/or removing identified words from the dynamic language model based on their respective time stamps. A screen scraper can capture text from the device display associated with the recorded utterance. The device display can contain customer service data.

Abstract: A method includes: loading, by a processor, a grammar specification defining at least one parameterizable grammar including a plurality of rules; setting, by the processor, an initial state of a grammar processor as a current state, the current state including parameters supplied to the rules; selecting, by the processor, a rule of the plurality of rules matching the parameters of the current state of the grammar processor; applying, by the processor, the selected rule to the audio and updating the current state; determining, by the processor, whether termination conditions have been met; in response to determining the termination conditions are not met, selecting, by the processor, from the plurality of rules in accordance with parameters of the updated state; and in response to determining the termination conditions are met, outputting, by the processor, a recognizer result of the current state.

Abstract: A text-to-speech (TTS) system that is capable of considering characteristics of various portions of text data in order to create continuity between segments of synthesized speech. The system can analyze text portions of a work and create feature vectors including data corresponding to characteristics of the individual portions and/or the overall work. A TTS processing component can then consider feature vector(s) from other portions when performing TTS processing on text of a first portion, thus giving the TTS component some intelligence regarding other portions of the work, which can then result in more continuity between synthesized speech segments.

Abstract: A machine translation system capable of clustering training data and performing dynamic domain adaptation is disclosed. An unsupervised domain clustering process is utilized to identify domains in general training data that can include in-domain training data and out-of-domain training data. Segments in the general training data are then assigned to the domains in order to create domain-specific training data. The domain-specific training data is then utilized to create domain-specific language models, domain-specific translation models, and domain-specific model weights for the domains. An input segment to be translated can be assigned to a domain at translation time. The domain-specific model weights for the assigned domain can be utilized to translate the input segment.

Abstract: Techniques for controlling a voice activated feature of a voice activated device are described. Data from one or more sensors and data indicative of a status of a user are received. Based on the analyzing the data, a proximity of the user relative to the location of the voice activated device is determined. One or more voice activated features of the voice activated device are enabled based at least in part on the determined proximity, one or more rules, and one or more user preferences.

Abstract: According to an embodiment, a recognition apparatus includes one or more processors. The one or more processors are configured to calculate, based on the input signal, a score vector sequence in which a plurality of score vectors each including respective scores of symbols are arranged; and cause, among: a first score vector in which a representative symbol corresponding to a best score is a recognition-target symbol; a second score vector in which a representative symbol is a non-target symbol, and a score of the representative symbol is worse than a first threshold; and a third score vector in which a representative symbol is a non-target symbol, and a score of the representative symbol is equal to the first threshold or better than the first threshold, a third score vector satisfying a predefined first condition, to pass through to filter the score vector sequence.

Abstract: Embodiments of the present invention disclose a transmission method, where the method includes: starting a session between a terminal and a peer terminal; detecting whether information transmitted in the session satisfies a trigger condition; and if the information satisfies the trigger condition, acquiring an image, and sending the image to the peer terminal, where the information transmitted in the session includes text information or voice information. According to the present invention, information transmission efficiency can be improved.

Abstract: Systems and methods for determining knowledge-guided information for a recurrent neural networks (RNN) to guide the RNN in semantic tagging of an input phrase are presented. A knowledge encoding module of a Knowledge-Guided Structural Attention Process (K-SAP) receives an input phrase and, in conjunction with additional sub-components or cooperative components generates a knowledge-guided vector that is provided with the input phrase to the RNN for linguistic semantic tagging. Generating the knowledge-guided vector comprises at least parsing the input phrase and generating a corresponding hierarchical linguistic structure comprising one or more discrete sub-structures. The sub-structures may be encoded into vectors along with attention weighting identifying those sub-structures that have greater importance in determining the semantic meaning of the input phrase.

Abstract: A method for controlling a drone includes receiving a request for information about a spatial location, generating data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: An electronic device and a voice recognition method thereof are provided. A voice recognition method of an electronic device includes receiving trigger voice, storing a characteristic of the trigger voice for voice verification by analyzing the trigger voice. When a user issues a voice command, determining whether the user voice command comprises the characteristic belonging to the trigger voice, and performing function of the electronic device corresponding to the user voice command based on the determination. Accordingly, users may easily use the voice recognition function of the electronic device.

Abstract: Various implementations disclosed herein include a phonotactic post-processor configured to rescore the N-best phoneme candidates output by a primary ensemble phoneme neural network using a priori phonotactic information. In various implementations, one of the scored set of the N-best phoneme candidates is selected as a preferred estimate for a one-phoneme output decision by the phonotactic post-processor. In some implementations, the one-phoneme output decision is an estimate of the most likely detected and recognized phoneme in a frame based on a function of posterior probabilities generated by an ensemble phoneme neural network, as well as phonotactic information and statistical performance characterizations incorporated by the phonotactic post-processor.

Abstract: Systems and processes are disclosed for virtual assistant request recognition using live usage data and data relating to future events. User requests that are received but not recognized can be used to generate candidate request templates. A count can be associated with each candidate request template and can be incremented each time a matching candidate request template is received. When a count reaches a threshold level, the corresponding candidate request template can be used to train a virtual assistant to recognize and respond to similar user requests in the future. In addition, data relating to future events can be mined to extract relevant information that can be used to populate both recognized user request templates and candidate user request templates. Populated user request templates (e.g., whole expected utterances) can then be used to recognize user requests and disambiguate user intent as future events become relevant.

Abstract: Systems and methods are provided for analyzing voice-based audio inputs. A voice-based audio input associated with a user (e.g., wherein the voice-based audio input is a prompt or a command) is received and measures of one or more features are extracted. One or more parameters are calculated based on the measures of the one or more features. The occurrence of one or more mistriggers is identified by inputting the one or more parameters into a predictive model. Further, systems and methods are provided for identifying human mental health states using mobile device data. Mobile device data (including sensor data) associated with a mobile device corresponding to a user is received. Measurements are derived from the mobile device data and input into a predictive model. The predictive model is executed and outputs probability values of one or more symptoms associated with the user.

Abstract: In at least one exemplary embodiment for automated document identification and language dictation recognition systems, the system comprises a database capable of receiving a plurality of verbal records, the verbal record comprising at least one identifier and at least one verbal feature and a processor operably coupled to the database, where the processor has and executes a software program. The processor being operational to identify a subset of the plurality of verbal records from the database, extract at least one verbal feature from the identified records, analyze the at least one verbal feature of the subset of the plurality of verbal records, process the subset of the plurality of records using the analyzed feature according to at least one reasoning approach, generate a processed verbal record using the processed subset of the plurality of records, and deliver the processed verbal record to a recipient.

Abstract: A complex video event classification, search and retrieval system can generate a semantic representation of a video or of segments within the video, based on one or more complex events that are depicted in the video, without the need for manual tagging. The system can use the semantic representations to, among other things, provide enhanced video search and retrieval capabilities.

Abstract: A speech recognition system interprets both spoken system commands as well as application commands. Users may speak commands to an open microphone of a computing device that may be interpreted by at least two speech recognizers operating simultaneously. The first speech recognizer interprets operating system commands and the second speech recognizer interprets application commands. The system commands may include at least opening and closing an application and the application commands may include at least a game command or navigation within a menu. A reserve word may be used to identify whether the command is for the operation system or application. A user's cadence may also indicate whether the speech is a global command or application command. A speech recognizer may include a natural language software component located in a remote computing device, such as in the so-called cloud.

Abstract: According to an embodiment, a voice processing device includes an interface system, a determining processor, and a predicting processor. The interface system configured to receive neutral voice data representing audio in a neutral voice of a user. The determining processor configured to determine a predictive parameter based at least in part on the neutral voice data. The predicting processor configured to predict a voice conversion model for converting the neutral voice of the speaker to a target voice using at least the predictive parameter.

Abstract: A method for using speech disfluencies detected in speech input to assist in interpreting the input is provided. The method includes providing access to a set of content items with metadata describing the content items, and receiving a speech input intended to identify a desired content item. The method further includes detecting a speech disfluency in the speech input and determining a measure of confidence of a user in a portion of the speech input following the speech disfluency. If the confidence measure is lower than a threshold value, the method includes determining an alternative query input based on replacing the portion of the speech input following the speech disfluency with another word or phrase. The method further includes selecting content items based on comparing the speech input, the alternative query input (when the confidence measure is low), and the metadata associated with the content items.

Abstract: An air conditioning system includes a collector that obtains user's voice data; a voice recognizer that generates character-string data on the basis of the voice data; a memory that stores therein a search table having a standard keyword that indicates the type of sensitivity to the environment inside a room and derivative keywords that indicate a same type of sensitivity as the standard keyword, while indicating different sensitivity levels; an analyzer that determines the type and level of user's sensitivity on the basis of the character-string data and the search table; and a controller that outputs a control signal to an adjustment device in an indoor device, the adjustment device adjusting the environment inside a room on the basis of the type and level of sensitivity, which are determined by the analyzer.

Abstract: In at least one exemplary embodiment for common phrase identification and language dictation recognition systems and methods for using the same, the system comprises a database capable of receiving a plurality of verbal records, the verbal record comprising at least one identifier and at least one verbal feature and a processor operably coupled to the database, where the processor has and executes a software program. The processor being operational to identify a subset of the plurality of verbal records from the database, extract at least one verbal feature from the identified records, analyze the at least one verbal feature of the subset of the plurality of verbal records, process the subset of the plurality of records using the analyzed feature according to at least one reasoning approach, generate a processed verbal record using the processed subset of the plurality of records, and deliver the processed verbal record to a recipient.

Abstract: A mechanism is provided for improving predicate parses (or logical representations of a passage) using semantic knowledge. In response to encountering an ambiguous decision point during a syntactic analysis of a portion of natural language content, a candidate meaning of the ambiguous decision point is generated. Characteristics of the ambiguous decision point are evaluated based on a semantic knowledge base to determine a semantic meaning associated with the ambiguous decision point. A determination is made as to whether the semantic meaning supports or refutes the candidate meaning. In response to determining that the semantic meaning refutes the candidate meaning, the candidate meaning of the ambiguous decision point is overridden based on the semantic meaning to include the semantic meaning as a final meaning for the ambiguous decision point. The portion of natural language content is then processed based on the final meaning for the ambiguous decision point.

Abstract: A method for controlling a drone includes receiving a natural language request for information about a spatial location, parsing the natural language request into data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: An exemplary computer system configured to train an ASR using the output from a TTS engine.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting follow-up queries to an initial transcription of an utterance. In some implementations, one or more follow-up queries that are pre-associated with a transcription of an initial utterance of a user are identified. A new or modified language model in which a respective probability associated with one or more of the follow-up queries is increased with respect to an initial language model is obtained. Subsequent audio data corresponding to a subsequent utterance of the user is then received. The subsequent audio data is processed using the new or modified language model to generate a transcription of the subsequent utterance. The transcription of the subsequent utterance is then provided for output to the user.

Abstract: According to an embodiment, a decoder searches a finite state transducer and outputs an output symbol string corresponding to a signal that is input or corresponding to a feature sequence of signal that is input. The decoder includes a token operating unit and a duplication eliminator. The token operating unit is configured to, every time the signal or the feature is input, propagate each of a plurality of tokens, which is assigned with a state of the head of a path being searched, according to the finite state transducer. The duplication eliminator is configured to eliminate duplication of two or more tokens which have same state assigned thereto and for which respective previously-passed transitions are assigned with same input symbol.

Abstract: Methods, systems and computer program products are disclosed for detecting patterns in a data stream that match multi-pattern rules. One embodiment of the invention provides a method of recognizing a specified group of patterns in a data stream. The method comprises identifying a rule for said specified group of patterns in the data stream, and using a first array of finite state machines to scan the data stream for at least some of the patterns in the specified group. For patterns in the specified group that are found in the data stream by the first array of finite state machines, pattern identifiers are sent to a second array of finite state machines. The second array of finite state machines determines if the specified group of patterns is in the data stream in accordance with the identified rule by, at least in part, using said pattern identifiers.

Abstract: Systems and techniques are provided for pruning a node from a possible nodes list for Hidden Markov Model with label transition node pruning. The node may be a label transition node. A frame may be at a predicted segmentation point in decoding input with the Hidden Markov Model. The node may be scored at the frame. The node may be pruned from the possible nodes list for the frame when score for the node is greater than the sum of a best score among nodes on the possible nodes list for the frame and a beam threshold minus a penalty term. A possible nodes list may be generated for a subsequent frame using label selection. A second node may be pruned from the possible nodes list for the subsequent frame with early pruning.

Abstract: Technology is disclosed that improves language coverage by selecting sentences to be used as training data for a language processing engine. The technology accomplishes the selection of a number of sentences by obtaining a group of sentences, computing a score for each sentence, sorting the sentences based on their scores, and selecting a number of sentences with the highest scores. The scores can be computed by dividing a sum of frequency values of unseen words (or n-grams) in the sentence by a length of the sentence. The frequency values can be based on posts in one or more particular domains, such as the public domain, the private domain, or other specialized domains.

Abstract: One or more embodiments present a script to a user in an interactive script environment. A digital representation of a manuscript is analyzed. This digital representation includes a set of roles and a set of information associated with each role in the set of roles. An active role in the set of roles that is associated with a given user is identified based on the analyzing. At least a portion of the manuscript is presented to the given user via a user interface. The portion includes at least a subset of information in the set of information. Information within the set of information that is associated with the active role is presented in a visually different manner than information within the set of information that is associated with a non-active role, which is a role that is associated with a user other than the given user.