Abstract: Disclosed herein includes a system, a method, and a device for localizing and classifying content in a data set. A device can provide a sequence of portions of a data set to a neural network to generate a plurality of activations. Each activation of the plurality of activations can include at least one value from a layer of the neural network. The device can apply an attention vector to each activation of the plurality of activations to generate a sequence of values. A normalization function can be applied to the sequence of values to generate a sequence of attention scores according to the sequence of values. The device can identify or localize one or more portions in the sequence of portions of the data based in part on the sequence of attention scores.

Abstract: Technologies pertaining to slot filling are described herein. A deep neural network, a recurrent neural network, and/or a spatio-temporally deep neural network are configured to assign labels to words in a word sequence set forth in natural language. At least one label is a semantic label that is assigned to at least one word in the word sequence.

Abstract: Various technologies described herein pertain to executing a mixed query to search a database retained in a data repository. The mixed query includes a regular expression, which is a pattern of elements, and a semantic constraint. The elements in the regular expression include a first wildcard, where the semantic constraint restricts a meaning of the first wildcard. Moreover, the elements in the regular expression include explicit lexical constraint(s) and/or disparate wildcard(s). For instance, semantic constraint(s) can restrict meaning(s) of the disparate wildcard(s). The mixed query is executed to retrieve results that match the pattern of the elements in the regular expression and satisfy the semantic constraint(s).

Abstract: Various technologies described herein pertain to executing a mixed query to search a database retained in a data repository. The mixed query includes a regular expression, which is a pattern of elements, and a semantic constraint. The elements in the regular expression include a first wildcard, where the semantic constraint restricts a meaning of the first wildcard. Moreover, the elements in the regular expression include explicit lexical constraint(s) and/or disparate wildcard(s). For instance, semantic constraint(s) can restrict meaning(s) of the disparate wildcard(s). The mixed query is executed to retrieve results that match the pattern of the elements in the regular expression and satisfy the semantic constraint(s).

Abstract: Various technologies described herein pertain to executing a mixed query to search a database retained in a data repository. The mixed query includes a regular expression, which is a pattern of elements, and a semantic constraint. The elements in the regular expression include a first wildcard, where the semantic constraint restricts a meaning of the first wildcard. Moreover, the elements in the regular expression include explicit lexical constraint(s) and/or disparate wildcard(s). For instance, semantic constraint(s) can restrict meaning(s) of the disparate wildcard(s). The mixed query is executed to retrieve results that match the pattern of the elements in the regular expression and satisfy the semantic constraint(s).

Abstract: Recurrent conditional random field (R-CRF) embodiments are described. In one embodiment, the R-CFR receives feature values corresponding to a sequence of words. Semantic labels for words in the sequence of words are then generated and each label is assigned to the appropriate one of the words in the sequence of words. The R-CRF used to accomplish these tasks includes a recurrent neural network (RNN) portion and a conditional random field (CRF) portion. The RNN portion receives feature values associated with a word in the sequence of words and outputs RNN activation layer activations data that is indicative of a semantic label. The CRF portion inputs the RNN activation layer activations data output from the RNN for one or more words in the sequence of words and outputs label data that is indicative of a separate semantic label that is to be assigned to each of the words.

Abstract: Recurrent conditional random field (R-CRF) embodiments are described. In one embodiment, the R-CFR receives feature values corresponding to a sequence of words. Semantic labels for words in the sequence of words are then generated and each label is assigned to the appropriate one of the words in the sequence of words. The R-CRF used to accomplish these tasks includes a recurrent neural network (RNN) portion and a conditional random field (CRF) portion. The RNN portion receives feature values associated with a word in the sequence of words and outputs RNN activation layer activations data that is indicative of a semantic label. The CRF portion inputs the RNN activation layer activations data output from the RNN for one or more words in the sequence of words and outputs label data that is indicative of a separate semantic label that is to be assigned to each of the words.

Abstract: Technologies pertaining to slot filling are described herein. A deep neural network, a recurrent neural network, and/or a spatio-temporally deep neural network are configured to assign labels to words in a word sequence set forth in natural language. At least one label is a semantic label that is assigned to at least one word in the word sequence.

Abstract: Various technologies described herein pertain to executing a mixed query to search a database retained in a data repository. The mixed query includes a regular expression, which is a pattern of elements, and a semantic constraint. The elements in the regular expression include a first wildcard, where the semantic constraint restricts a meaning of the first wildcard. Moreover, the elements in the regular expression include explicit lexical constraint(s) and/or disparate wildcard(s). For instance, semantic constraint(s) can restrict meaning(s) of the disparate wildcard(s). The mixed query is executed to retrieve results that match the pattern of the elements in the regular expression and satisfy the semantic constraint(s).

Abstract: A computer-implemented speech recognition system described herein includes a receiver component that receives a plurality of detected units of an audio signal, wherein the audio signal comprises a speech utterance of an individual. A selector component selects a subset of the plurality of detected units that correspond to a particular time-span. A generator component generates at least one feature with respect to the particular time-span, wherein the at least one feature is one of an existence feature, an expectation feature, or an edit distance feature. Additionally, a statistical speech recognition model outputs at least one word that corresponds to the particular time-span based at least in part upon the at least one feature generated by the feature generator component.

Abstract: A computer-implemented speech recognition system described herein includes a receiver component that receives a plurality of detected units of an audio signal, wherein the audio signal comprises a speech utterance of an individual. A selector component selects a subset of the plurality of detected units that correspond to a particular time-span. A generator component generates at least one feature with respect to the particular time-span, wherein the at least one feature is one of an existence feature, an expectation feature, or an edit distance feature. Additionally, a statistical speech recognition model outputs at least one word that corresponds to the particular time-span based at least in part upon the at least one feature generated by the feature generator component.

Abstract: A voice interaction system is configured to analyze an utterance and identify inherent attributes that are indicative of a demographic characteristic of the system user that spoke the utterance. The system then selects and presents a personalized response to the user, the response being selected based at least in part on the identified demographic characteristic. In one embodiment, the demographic characteristic is one or more of the caller's age, gender, ethnicity, education level, emotional state, health status and geographic group. In another embodiment, the selection of the response is further based on consideration of corroborative caller data.

Abstract: A method of communicating information about a product evaluation between a system having a data store and a wireless client device is discussed. The method includes receiving a signal representative of an audible indication from the client device via a wireless communication link identifying the product about which evaluation information is to be communicated. The method further includes comparing an indication of the signal to data in the data store in response to match the indication with a portion of the data and communicating evaluation information between the wireless client device and the system.