Abstract: A computer system for narrating a table using at least one narration template, wherein the table is extracted from a data source is provided. The computer system may include parsing the extracted table. The computer system may also include performing structural analysis on the parsed extracted table. The computer system may further include selecting at least one structural template based on the structural analysis of the parsed extracted table. Additionally, the computer system may include selecting the at least one narration template based on the at least one selected structural template. The computer system may also include applying the at least one selected narration template to the extracted table. The computer system may further include narrating the extracted table based on the applying of the at least one selected narration template to the extracted table.

Abstract: A computer-implemented method for language generation of a flow diagram, which receives a flow diagram. A plurality of geometric shapes within the flow diagram is identified. A plurality of text elements within the flow diagram is identified. The plurality of text elements and corresponding geometric shapes are associated. The association between the plurality of geometric shapes are identified. A diagram matrix based on the associations between the plurality of geometric shapes is generated. A linear language representation of the diagram matrix is generated.

Abstract: A system and computer implemented method for managing perspective data is disclosed. The method may include collecting a first lot of perspective data for an item. The method may include introducing a variant feature to the item to constitute a modified item. The method may include collecting a second lot of perspective data for the modified item. The method may also include evaluating the first and second lots of perspective data to ascertain a sentiment fluctuation based on information relevant to the variant feature.

Abstract: A method for narrating a table using at least one narration template, wherein the table is extracted from a data source is provided. The method may include parsing the extracted table. The method may also include performing structural analysis on the parsed extracted table. The method may further include selecting at least one structural template based on the structural analysis of the parsed extracted table. Additionally, the method may include selecting the at least one narration template based on the at least one selected structural template. The method may also include applying the at least one selected narration template to the extracted table. The method may further include narrating the extracted table based on the applying of the at least one selected narration template to the extracted table.

Abstract: A method and system is provided for estimating clean speech parameters from noisy speech parameters. The method is performed by acquiring speech signals, estimating noise from the acquired speech signals, computing speech features from the acquired speech signals, estimating model parameters from the computed speech features and estimating clean parameters from the estimated noise and the estimated model parameters.

Abstract: According to one embodiment, a machine translation apparatus includes the following elements. The machine translation unit performs machine translation on a first text in a first language to generate a first machine translation result in a second language. The retrieval unit retrieves a first question sentence in the first language similar to the first text to obtain a degree of similarity between the first text and the first question sentence. The determination unit determines a first answer sentence in the first language corresponding to the first question sentence to be an output target when the degree of similarity is higher than a threshold and determines the first machine translation result to be an output target when the degree of similarity is lower than the threshold.

Abstract: There is provided a system including a microphone configured to receive an input speech, an analog to digital (A/D) converter configured to convert the input speech to a digital form and generate a digitized speech including a plurality of segments having acoustic features, a memory storing an executable code, and a processor executing the executable code to extract a plurality of acoustic feature vectors from a first segment of the digitized speech, determine, based on the plurality of acoustic feature vectors, a plurality of probability distribution vectors corresponding to the probabilities that the first segment includes each of a first keyword, a second keyword, both the first keyword and the second keyword, a background, and a social speech, and assign a first classification label to the first segment based on an analysis of the plurality of probability distribution vectors of one or more segments preceding the first segment and the probability distribution vectors of the first segment.

Abstract: In an approach for selectively displaying a push notification, audio is captured using a microphone. A processor receives a push notification, wherein the push notification includes information. A processor identifies a keyword associated with the push notification based on the information. A processor determines that the captured audio includes the keyword. A processor determines whether to display the push notification based on the determination of whether the captured audio includes the keyword.

Abstract: An encoding device according to the disclosure includes a first encoding unit that generates a first encoded signal in which a low-band signal having a frequency lower than or equal to a predetermined frequency from a voice or audio input signal is encoded, and a low-band decoded signal; a second encoding unit that encodes, on the basis of the low-band decoded signal, a high-band signal having a band higher than that of the low-band signal to generate a high-band encoded signal; and a first multiplexing unit that multiplexes the first encoded signal and the high-band encoded signal to generate and output an encoded signal. The second encoding unit calculates an energy ratio between a high-band noise component, which is a noise component of the high-band signal, and a high-band non-tonal component of a high-band decoded signal generated from the low-band decoded signal and outputs the ratio as the high-band encoded signal.

Abstract: A system and method are presented for forming the excitation signal for a glottal pulse model based parametric speech synthesis system. The excitation signal may be formed by using a plurality of sub-band templates instead of a single one. The plurality of sub-band templates may be combined to form the excitation signal wherein the proportion in which the templates are added is dynamically based on determined energy coefficients. These coefficients vary from frame to frame and are learned, along with the spectral parameters, during feature training. The coefficients are appended to the feature vector, which comprises spectral parameters and is modeled using HMMs, and the excitation signal is determined.

Abstract: The techniques described herein improve methods to equip a computing device to conduct automatic speech recognition (“ASR”) in talker-independent multi-talker scenarios. In some examples, permutation invariant training of deep learning models can be used for talker-independent multi-talker scenarios. In some examples, the techniques can determine a permutation-considered assignment between a model's estimate of a source signal and the source signal. In some examples, the techniques can include training the model generating the estimate to minimize a deviation of the permutation-considered assignment. These techniques can be implemented into a neural network's structure itself, solving the label permutation problem that prevented making progress on deep learning based techniques for speech separation. The techniques discussed herein can also include source tracing to trace streams originating from a same source through the frames of a mixed signal.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for invoking an agent module in an automated assistant application in response to user selection of a selectable element presented at a graphical user interface rendered by a non-automated assistant application. The invoked agent module can be associated with other content rendered in the non-automated assistant graphical user interface, and can optionally be invoked with values that are based on user interactions via the non-automated assistant application. Responsive content can be received from the agent module in response to the invocation, and corresponding content provided by the automated assistant application via an automated assistant interface. In these and other manners, selection of the selectable element causes transition from a non-conversational interface, to a conversational automated assistant interface—where an agent (relevant to content in the non-conversational interface) is invoked in the automated assistant interface.

Abstract: An audio/speech encoding apparatus/method and an audio/speech decoding apparatus/method are provided. The audio/speech encoding apparatus includes a memory that stores instructions, and a processor that performs operations. The operations include transforming a time domain input audio/speech signal to a frequency spectrum, dividing the frequency spectrum to a plural of bands, calculating norm factors, and quantizing the norm factors. The operations also include calculating differential indices between an Nth band index and an (N?1)th band index, and modifying a range of the differential indices for the Nth band when N is 2 or more. The operations further include replacing the differential index with the modified differential index, and not modifying a range of the differential indices for the Nth band when N is 1. The apparatus encodes the differential indices using a selected Huffman table, and transmits the encoded differential indices and a flag signal over a communication network.

Abstract: Systems, methods, and computer-readable storage devices for performing per-channel automatic speech recognition. An example system configured to practice the method combines a first audio signal of a first speaker in a communication session and a second audio signal from a second speaker in the communication session as a first audio channel and a second audio channel. The system can recognize speech in the first audio channel of the recording using a first model specific to the first speaker, and recognize speech in the second audio channel of the recording using a second model specific to the second speaker, wherein the first model is different from the second model. The system can generate recognized speech as an output from the communication session. The system can identify the models based on identifiers of the speakers, such as a telephone number, an IP address, a customer number, or account number.

Abstract: Embodiments described herein provide approaches for validating synonyms in ontology driven natural language processing. Specifically, an approach is provided for receiving a user input containing a token, structuring the user input into a semantic model comprising a set of classes each containing a set of related permutations of the token, designating the token as a synonym of one of the set of related permutations, annotating the token with a class from the set of classes corresponding to the one of the set of related permutations, and validating the annotation of the token by determining an accuracy of the designation of the token as a synonym of the one of the set of related permutations. In one embodiment, the accuracy is determined by quantifying a linear distance between the token and a contextual token also within the user input, and comparing the linear distance to a pre-specified linear distance limit.

Abstract: An electronic device or method for adjusting a gain on a voice operated control system can include one or more processors and a memory having computer instructions. The instructions, when executed by the one or more processors causes the one or more processors to perform the operations of receiving a first microphone signal, receiving a second microphone signal, updating a slow time weighted ratio of the filtered first and second signals, and updating a fast time weighted ratio of the filtered first and second signals. The one or more processors can further perform the operations of calculating an absolute difference between the fast time weighted ratio and the slow time weighted ratio, comparing the absolute difference with a threshold, and increasing the gain when the absolute difference is greater than the threshold. Other embodiments are disclosed.

Abstract: Features are disclosed for managing the use of speech recognition models and data in automated speech recognition systems. Models and data may be retrieved asynchronously and used as they are received or after an utterance is initially processed with more general or different models. Once received, the models and statistics can be cached. Statistics needed to update models and data may also be retrieved asynchronously so that it may be used to update the models and data as it becomes available. The updated models and data may be immediately used to re-process an utterance, or saved for use in processing subsequently received utterances. User interactions with the automated speech recognition system may be tracked in order to predict when a user is likely to utilize the system. Models and data may be pre-cached based on such predictions.

Abstract: A method includes converting a user's utterance to text; encapsulating the converted text in a rheme object; searching, for each of a plurality of topics, for keywords in the converted text; determining a relevancy metric for each of the plurality of topics based on such searching; selecting one or more topics based on determined relevancy metrics; comparing some or all of the converted text to names in one or more patient lists or databases; identifying a unique patient whose name is contained in the converted; attaching an indication of the identified patient to the rheme object; effecting an action based on the selected one or more topics and the attached patient indication; and saving the topic in a conversation history with a reference to the identified patient.

Abstract: The invention provides a decoder being configured for processing an encoded audio bitstream, wherein the decoder includes: a bitstream decoder configured to derive a decoded audio signal from the bitstream, wherein the decoded audio signal includes at least one decoded frame; a noise estimation device configured to produce a noise estimation signal containing an estimation of the level and/or the spectral shape of a noise in the decoded audio signal; a comfort noise generating device configured to derive a comfort noise signal from the noise estimation signal; and a combiner configured to combine the decoded frame of the decoded audio signal and the comfort noise signal in order to obtain an audio output signal.

Abstract: Embodiments relate to censoring audio data. A censoring system receives audio data including a song tag and amplitude data as a function of time. The amplitude data represents spoken words occurring over a duration, as well as non-spoken word sound overlapping with some of the spoken words during the duration. The system accesses a set of song lyrics and processes the set of song lyrics and the amplitude data together to identify timestamps in the amplitude data. These timestamps indicate a time during the duration when one of the words from the lyrics begins in the amplitude data. The system compares the words in the set of song lyrics to a blacklist and adjusts the amplitude data at the timestamps of blacklisted word occurrences to render the audio at the blacklisted words incomprehensible. The system outputs the adjusted amplitude data.