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: A system for detecting an original language of a translated document retrieves the translated document, and identifies a language of the retrieved document. The system calculates a language model for the language of the retrieved document (LM(RD)). The system calculates a distinct vector as a difference between LM(RD) and a common language model for the language of the retrieved document (LMT(RD)). The system obtains pair vectors for language model pairs associated with the language of the retrieved document, and calculates a vector distance between the distinct vector and each pair vector (or between the (LM(RD)) and each pair vector). The system identifies a given pair vector within a threshold vector distance, and calculates the confidence score. The system then identifies the original language corresponding to the given pair vector as the original language of the retrieved document, and retrieves an original document in the original language of the retrieved document.

Abstract: Systems and processes for converting speech-to-text are provided. In one example process, speech input can be received. A sequence of states and arcs of a weighted finite state transducer (WFST) can be traversed. A negating finite state transducer (FST) can be traversed. A virtual FST can be composed using a neural network language model and based on the sequence of states and arcs of the WFST. The one or more virtual states of the virtual FST can be traversed to determine a probability of a candidate word given one or more history candidate words. Text corresponding to the speech input can be determined based on the probability of the candidate word given the one or more history candidate words. An output can be provided based on the text corresponding to the speech input.

Abstract: Disclosed is a speaker diarization process for determining which speaker is speaking at what time during the course of a conversation. The entire process can be most easily described in five main parts: Segmentation where speech/non-speech decisions are made; frame feature extraction where useful information is obtained from the frames; segment modeling where the information from the frame feature extraction is combined with segment start and end time information to create segment specific features; speaker decisions when the segments are clustered to create speaker models; and corrections where frame level corrections are applied to the information extracted.

Abstract: A method is presented for forming the excitation signal for a glottal pulse model based parametric speech synthesis system. In one embodiment, fundamental frequency values are used to form the excitation signal. The excitation is modeled using a voice source pulse selected from a database of a given speaker. The voice source signal is segmented into glottal segments, which are used in vector representation to identify the glottal pulse used for formation of the excitation signal. Use of a novel distance metric and preserving the original signals extracted from the speakers voice samples helps capture low frequency information of the excitation signal. In addition, segment edge artifacts are removed by applying a unique segment joining method to improve the quality of synthetic speech while creating a true representation of the voice quality of a speaker.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for assigning saliency weights to words of an ASR model. The saliency values assigned to words within an ASR model are based on human perception judgments of previous transcripts. These saliency values are applied as weights to modify an ASR model such that the results of the weighted ASR model in converting a spoken document to a transcript provide a more accurate and useful transcription to the user.

Abstract: Obtaining and/or validating user credentials at client devices is described. A phrase may be generated based on one or more index values determined according to a function of time and a credential identifier identifying a user credential. The phrase may be output by the client device for validating the user credential.

Abstract: In general, techniques are described for determining quantization step sizes for compression of spatial components of a sound field. A device comprising one or more processors may be configured to perform the techniques. In other words, the one or more processors may be configured to determine a quantization step size to be used when compressing a spatial component of a sound field, where the spatial component generated by performing a vector based synthesis with respect to a plurality of spherical harmonic coefficients.

Abstract: A voice detection method which makes it possible to detect the presence of voice signals in an noisy acoustic signal x(t) from a microphone, including the following consecutive steps: calculating a detection function FD(?) based on calculating a difference function D(?) varying in accordance with the shift ? on an integration window with length W starting at the time t0, with: a step of adapting the threshold in said current interval, in accordance with values calculated from the acoustic signal x(t) established in said current interval; searching for the minimum of the detection function FD(?) and comparing the minimum with a threshold, for (?) varying in a predetermined time interval referred to as current interval so as to detect the possible presence of a fundamental frequency F0 that is characteristic of a voice signal in said current interval.

Abstract: An image processing device, comprises: an input part for inputting image data; a word extracting part for extracting a word from texts contained in the image data; a synonym obtaining part for obtaining a synonym corresponds to the word, and for associating the obtained synonym with the word; a position identifying part for identifying a display position on the image data of the word with which the synonym is associated; a layer creating part for creating an accompanying layer to add to an original layer, which is the image data containing the word, and for embedding the synonym associated with the word within a position on the accompanying layer the same as the display position identified by the position identifying part; and an output image generating part for generating output image data including the original layer containing the word and the accompanying layer within which the synonym is embedded.

Abstract: A system and method for navigating secondary content. The system may monitor for gestures input to the system by an input device and may detect an arc gesture. The arc gesture may travel along both a horizontal axis and a vertical axis from a first point to a second point and may be delineated from a horizontal or a vertical motion. The system may identify secondary content corresponding to the arc gesture in response to the arc gesture and output data corresponding to the secondary content. The system may identify supplemental text associated with the secondary content and synthesize supplemental speech corresponding to the supplemental text. The output data may include audio including the synthesized supplemental speech.

Abstract: A method includes providing a deep neural network acoustic model, receiving audio data including one or more utterances of a speaker, extracting a plurality of speech recognition features from the one or more utterances of the speaker, creating a speaker identity vector for the speaker based on the extracted speech recognition features, and adapting the deep neural network acoustic model for automatic speech recognition using the extracted speech recognition features and the speaker identity vector.

Abstract: Mechanisms are provided for performing tabular data correction in a document. The mechanisms receive a natural language document comprising a portion of content and analyze the portion of content within the natural language document to identify an erroneous sub-portion comprising an erroneous or missing item of information. The mechanisms generate a semantic signature for the erroneous sub-portion and generate a query based on the semantic signature. The mechanisms apply the query to a knowledge base to identify a candidate sub-portion of content. The mechanisms correct the erroneous sub-portion using the identified candidate sub-portion of content to generate a corrected natural language document.

Abstract: Methods and systems are provided for receiving desired sounds. The system includes a position sensor configured to determine an occupant position of an occupant engaging in speech within a defined space and transmit the speaking occupant position. A plurality of microphones are configured to receive sound from within the defined space and transmit audio signals corresponding to the received sound. A processor, in communication with the position sensor and the microphones, is configured to receive the speaking occupant position and the audio signals, apply a beamformer to the audio signals to direct a microphone beam toward the occupant position, and generate a beamformer output signal.

Abstract: A system, method and computer-readable storage device which balance latency and accuracy of machine translations by segmenting the speech upon locating a conjunction. The system, upon receiving speech, will buffer speech until a conjunction is detected. Upon detecting a conjunction, the speech received until that point is segmented. The system then continues performing speech recognition on the segment, searching for the next conjunction, while simultaneously initiating translation of the segment. Upon translating the segment, the system converts the translation to a speech output, allowing a user to hear an audible translation of the speech originally heard.

Abstract: A method estimates source signals from a mixture of source signals by first training an analysis model and a reconstruction model using training data. The analysis model is applied to the mixture of source signals to obtain an analysis representation of the mixture of source signals, and the reconstruction model is applied to the analysis representation to obtain an estimate of the source signals, wherein the analysis model utilizes an analysis linear basis representation, and the reconstruction model utilizes a reconstruction linear basis representation.

Abstract: A device for aiding communication in the aeronautical domain, wherein the device includes a transceiver and data processor assembly that records audio messages corresponding to all the incoming and outgoing audio communications, transcribes the messages, in real time, into textual messages, displays the textual messages, and enables an audio play back of the audio messages.

Abstract: Disclosed herein are systems, methods, and computer-readable storage devices for processing audio signals. An example system configured to practice the method receives audio at a device to be transmitted to a remote speech processing system. The system analyzes one of noise conditions, need for an enhanced speech quality, and network load to yield an analysis. Based on the analysis, the system determines to bypass user-defined options for enhancing audio for speech processing. Then, based on the analysis, the system can modify an audio transmission parameter used to transmit the audio from the device to the remote speech processing system. The audio transmission parameter can be one of an amount of coding, a chosen codec, an amount of coding, or a number of audio channels, for example.

Abstract: A system and computer product for validating the consistency between quantitative and natural language textual evaluations. An example method involves computing a numeric score for a textual evaluation, comparing the numeric score to a quantitative evaluation, and producing a rating based on the similarity of the two evaluations.

Abstract: A system and computer product for validating the consistency between quantitative and natural language textual evaluations. An example method involves computing a numeric score for a textual evaluation, comparing the numeric score to a quantitative evaluation, and producing a rating based on the similarity of the two evaluations.