Abstract: An apparatus for notification of speech of interest to a user includes a voice analyzer configured to recognize speech, evaluate a relevance between a result of the speech recognition and a determined user's topic of interest, and determine whether to provide a notification; and an outputter configured to, in response to the voice analyzer determining to provide the notification, generate and output a notification message.

Abstract: Disclosed methods and systems are directed to concept identification and capture. The methods and systems may include receiving, by a device, a first natural language input comprising one or more terms, and analyzing the first natural language input via a natural language processing engine to identify one or more named entities associated with the one or more terms, wherein each of the one or more named entities is associated with at least one category of a plurality of categories. The methods and systems may also include detecting a text field configured to receive text, the text field being associated with one of the plurality of categories, and inputting into the text field one of the one or more identified named entities based on the text field being associated with a same category as the one of the one or more named entities.

Abstract: An approach to speech synthesis uses two phases in which a relatively low quality waveform is computed, and that waveform is passed through an enhancement phase which generates the waveform that is ultimately used to produce the acoustic signal provided to the user. For example, the first phase and the second phase are each implemented using a separate artificial neural network. The two phases may be computationally preferable to using a direct approach to yield a synthesized waveform of comparable quality.

Abstract: A processing result error detection device includes: a behavior recognition unit which recognizes behavior of a user; a processing unit which executes a process dependent on the behavior of the user; an emotion presumption unit which presumes an intensity of emotion of the user based on reaction of the user to a processing result; and a judgment unit which judges that the processing result includes an error when the presumed intensity of emotion of the user is equal to or greater than a predetermined threshold.

Abstract: A personal audio device, such as a wireless telephone, includes noise canceling circuit that adaptively generates an anti-noise signal from a reference microphone signal and injects the anti-noise signal into the speaker or other transducer output to cause cancellation of ambient audio sounds. An error microphone may also be provided proximate the speaker to estimate an electro-acoustical path from the noise canceling circuit through the transducer. A processing circuit uses the reference and/or error microphone, optionally along with a microphone provided for capturing near-end speech, to determine whether one of the reference or error microphones is obstructed by comparing their received signal content and takes action to avoid generation of erroneous anti-noise.

Abstract: A method comprising of receiving a first communication content directed to a user. The first communication content includes one or a combination of the following: content read by the user and content written by the user. The method also comprises of generating tokens corresponding to the first communication content by applying natural language processing and generating a token frequency index for the user, based on the tokens generated from the first communication content. The method determines a lexicon reading level for the user, based on the token frequency index generated for the user. The lexicon reading level indicates a reading level of the user. The method adds the lexicon reading level to a lexicon profile of the user. The method modifies a second communication content by replacing tokens with synonyms of the tokens based on comparing the difficulty ratings of the tokens with the user's lexicon reading level.

Abstract: An audio coding method and apparatus, where the method includes, for each audio frame in audio, when a signal characteristic of the audio frame and a signal characteristic of a previous audio frame meet a preset modification condition, determining a first modification weight according to linear spectral frequency (LSF) differences of the audio frame and the LSF differences of the previous audio frame, modifying a linear predictive parameter of the audio frame according to the determined first modification weight, and coding the audio frame according to a modified linear predictive parameter of the audio frame. According to the present disclosure, audio having a wider bandwidth can be coded while a bit rate remains unchanged or a bit rate slightly changes and a spectrum between audio frames is steadier.

Abstract: A system and method that translates sentences of natural language text into sets of axioms of formal logic that are consistent with parses resulting from NLP and acquired constraints as they accumulate. The system and method further present these axioms so as to facilitate further disambiguation of such sentences and produces axioms of formal logic suitable for processing by automated reasoning technologies, such as first-order or description logic suitable for processing by various reasoning algorithms, such as logic programs, inference engines, theorem provers, and rule-based systems.

Abstract: A collaborative recognition apparatus, the recognition apparatus includes a recognition data generator configured to generate recognition data based on sensor data, a processor configured to calculate reliability of the recognition data and to determine whether collaboration is desired based on the calculated reliability, and request collaboration data from the external recognition apparatus, and correct the recognition data based on the collaboration data received from the external recognition apparatus to generate final recognition data, in response to a determination that the collaboration is desired.

Abstract: There is provided an information processing device including an analysis unit configured to analyze a character string indicating contents of utterance obtained as a result of speech recognition, and a display control unit configured to display the character string indicating the contents of the utterance and an analysis result on a display screen.

Abstract: A computer implemented method for automatically moderating a system response to a user's voice command. A voice command from a user is received. The voice command is associated with a system command, the system command including command requirements. A determination is the made as to whether the user is experiencing stress based on a stress level detected in the received voice command, and the command requirements dynamically adjusted when the user is determined to be experiencing stress.

Abstract: An electronic device comprising a processor which is configured to perform speech recognition on an audio signal, linguistically analyze the output of the speech recognition for named-entities, perform an Internet or database search for the recognized named-entities to obtain query results, and display, on a display of the electronic device, information obtained from the query results on a timeline.

Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for utilizing multiple AI service providers by a dialog management system. The dialog management system can include a dispatcher bot, multiple worker bots, and multiple AI adapters that are each associated with a different cloud-based AI service provider. In response to receiving a query, the dispatcher bot selects a particular worker bot to handle the query. The particular worker bot is assigned to a particular AI service provider. An AI adapter associated with the particular AI service provider, generates a query message based on the query. The AI adapter sends the query message to the particular AI service provider and receives a response message. The dialog management system sends a representation of the response message to the particular worker bot, receives an answer for the query from the particular worker bot, and provides the answer for output.

Abstract: Described herein are systems and methods for automatic unit selection and target decomposition for sequence labelling. Embodiments include a new loss function called Gram-Connectionist Temporal Classification (CTC) loss that extend the popular CTC loss function criterion to alleviate prior limitations. While preserving the advantages of CTC, Gram-CTC automatically learns the best set of basic units (grams), as well as the most suitable decomposition of target sequences. Unlike CTC, embodiments of Gram-CTC allow a model to output variable number of characters at each time step, which enables the model to capture longer term dependency and improves the computational efficiency. It is also demonstrated that embodiments of Gram-CTC improve CTC in terms of both performance and efficiency on the large vocabulary speech recognition task at multiple scales of data, and that systems that employ an embodiment of Gram-CTC can outperform the state-of-the-art on a standard speech benchmark.

Abstract: Disclosed herein are methods of diarizing audio data using first-pass blind diarization and second-pass blind diarization that generate speaker statistical models, wherein the first pass-blind diarization is on a per-frame basis and the second pass-blind diarization is on a per-word basis, and methods of creating acoustic signatures for a common speaker based only on the statistical models of the speakers in each audio session.

Abstract: A system for identifying address components includes an interface and a processor. The interface is to receive an address for parsing. The processor is to determine a matching model of a set of models based at least in part on a matching probability for each model for a tokenized address, which is based on the address for parsing, and associate each component of the tokenized address with an identifier based at least in part on the matching model, wherein each component of the set of components is associated with an identifier, and wherein probabilities of each component of the set of components are determined using training addresses.

Abstract: A home assistant device assisting with the setup and troubleshooting of devices is described. An assistant device can determine setup instructions for devices. The setup instructions for one device can be determined the technical aptitude of the user and assistant the user with setting up devices within their own technical abilities. The setup instructions can then be provided.

Abstract: There is provided a method and a system for generating a summary of digital content. The method comprises: executing a syntax analysis of a textual representation of the digital content; segmenting the digital content into an ordered set of fragments (i.e. a first fragment and a second fragment); executing a semantic analysis of each fragment of the textual representation; determining a utility parameter for each fragment of the set of fragments; determining a linkage between each pair of fragments of the set of fragments; in response to the utility parameter of the second fragment exceeding a pre-determined threshold value, including the second fragment in a subset of fragments for inclusion in the summary of the digital content; in response to the linkage having been determined between the second fragment and the first fragment, including the first fragment in the subset of fragments; and generating the summary of the digital content.

Abstract: A likely source language of a media item can be identified by attempting an initial language identification of the media item based on intrinsic or extrinsic factors, such as words in the media item and languages known by the media item author. This initial identification can generate a list of most likely source languages with corresponding likelihood factors. Translations can then be performed presuming each of the most likely source languages. The translations can be performed for multiple output languages. Each resulting translation can receive a corresponding score based on a number of factors. The scores can be combined where they have a common source language. These combined scores can be used to weight the previously identified likelihood factors for the source languages of the media item.

Abstract: An embodiment of a speech endpoint detector apparatus may include a speech detector to detect a presence of speech in an electronic speech signal, a pause duration measurer communicatively coupled to the speech detector to measure a duration of a pause following a period of detected speech, an end of utterance detector communicatively coupled to the pause duration measurer to detect if the pause measured following the period of detected speech is greater than a pause threshold corresponding to an end of an utterance, and a pause threshold adjuster to adaptively adjust the pause threshold corresponding to an end of an utterance based on stored pause information. Other embodiments are disclosed and claimed.