Abstract: Systems and methods of providing text related to utterances, and gathering voice data in response to the text are provide herein. In various implementations, an identification token that identifies a first file for a voice data collection campaign, and a second file for a session script may be received from a natural language processing training device. The first file and the second file may be used to configure the mobile application to display a sequence of screens, each of the sequence of screens containing text of at least one utterance specified in the voice data collection campaign. Voice data may be received from the natural language processing training device in response to user interaction with the text of the at least one utterance. The voice data and the text may be stored in a transcription library.

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: Provided are a method and apparatus for encoding/decoding stereo audio. In the method for encoding stereo audio, stereo audio is encoded based on at least one of the phase difference between first and second channel audios and information on an angle made by a vector on the intensity of mono-audio and a vector on the intensity of the first channel audio or a vector on the intensity of the second channel audio. Thus, the number of encoded parameters is minimized so that a compression ratio in the encoding of the stereo audio is improved.

Abstract: There is provided an audio encoding apparatus that can avoid that audio data becomes irreproducible after fast-forward play. A quantization unit quantizes and buffers audio data into a buffer unit. A stream generating unit puts buffered audio data in a frame where there is a header related to the audio data in a stream and/or in one or plural frames preceding that frame. As for a predetermined frame, the stream generating unit puts in a data field of the frame the whole of an audio data piece related to a header included in that frame and puts audio sample data following that audio sample in a remaining part of the data field. As for a frame not a predetermined one, it puts in a data field of the frame an audio data piece related to a header included in that frame and/or audio data pieces following that audio data piece.

Abstract: An additive three-dimensional fabrication system includes voice control for user interaction. This voice-controlled interface can enable a variety of voice-controlled functions and operations, while supporting interactions specific to consumer-oriented fabrication processes.

Abstract: Systems and methods for managing an emergency situation are provided herein. According to some embodiments, the present technology may related to a security system and method for monitoring, detecting, and providing notification and/or response measures in response to an emergency situation regarding a user.

Abstract: The present invention includes systems and methods for sending social media messages without the need for keyboard inputs. A microphone captures live audio speech data and transmits the audio data to a processing unit. The processing unit converts the audio to speech data. The processing unit also removes censored words, emphasizes key words, and edits that data to include product and promotional messages where appropriate. The processing unit then uses code words contained in the speech data to send the speech data to the appropriate social media outlets for output.

Abstract: Disclosed herein is a character input apparatus including: a display section having a screen capable of displaying at least characters; an operation section configured to allow a user to input at least the characters; a first character input processing section configured to perform a first character input process of causing a character string to be displayed on the screen in accordance with a predetermined notation rule; a second character input processing section configured to perform a second character input process of causing a character string to be displayed on the screen not in accordance with the predetermined notation rule; a scene determination section configured to determine a character input scene; and an input process switch control section configured to switch between the first character input process and the second character input process in accordance with the character input scene.

Abstract: Disclosed herein are systems, methods, and computer-readable storage media for selecting a speech recognition model in a standardized speech recognition infrastructure. The system receives speech from a user, and if a user-specific supervised speech model associated with the user is available, retrieves the supervised speech model. If the user-specific supervised speech model is unavailable and if an unsupervised speech model is available, the system retrieves the unsupervised speech model. If the user-specific supervised speech model and the unsupervised speech model are unavailable, the system retrieves a generic speech model associated with the user. Next the system recognizes the received speech from the user with the retrieved model. In one embodiment, the system trains a speech recognition model in a standardized speech recognition infrastructure. In another embodiment, the system handshakes with a remote application in a standardized speech recognition infrastructure.

Abstract: A speaker intent analysis system and method for validating the truthfulness and intent of a plurality of participants' responses to questions. A computer stores, retrieves, and transmits a series of questions to be answered audibly by participants. The participants' answers are received by a data processor. The data processor analyzes and records the participants' speech parameters for determining the likelihood of dishonesty. In addition to analyzing participants' speech parameters for distinguishing stress or other abnormality, the processor may be equipped with voice recognition software to screen responses that while not dishonest, are indicative of possible malfeasance on the part of the participants. Once the responses are analyzed, the processor produces an output that is indicative of the participant's credibility. The output may be sent to proper parties and/or devices such as a web page, computer, e-mail, PDA, pager, database, report, etc. for appropriate action.

Abstract: In one embodiment, the invention provides a method for machine translation of a source document in an input language to a target document in an output language, comprising generating translation options corresponding to at least portions of each sentence in the input language; and selecting a translation option for the sentence based on statistics associated with the translation options.

Abstract: A voice agent device includes: a position detection unit which detects a position of a person in a conversation space to which the voice agent device is capable of providing information; a voice volume detection unit which detects a voice volume of the person from a sound signal in the conversation space obtained by a sound acquisition unit; a conversation area determination unit which determines a conversation area as a first area including the position when the voice volume has a first voice volume value and determines the conversation area as a second area including the position and being smaller than the first area when the voice volume has a second voice volume value smaller than the first voice volume value, the conversation area being a spatial range where an utterance of the person can be heard; and an information provision unit which provides provision information to the conversation area.

Abstract: In some embodiments, a multistage filter whose biquad filter stages are combined with latency between the stages, a system (e.g., an audio encoder or decoder) including such a filter, and methods for multistage biquad filtering. In typical embodiments, all biquad filter stages of the filter are operable independently to perform fully parallelized processing of data. In some embodiments, the inventive multistage filter includes a buffer memory, at least two biquad filter stages, and a controller coupled and configured to assert a single stream of instructions to the filter stages. Typically, the multistage filter is configured to perform multistage filtering of a block of input samples in a single processing loop with iteration over a sample index but without iteration over a biquadratic filter stage index.

Abstract: Method, system and medium for character converting between different regional versions of a language especially between Simplified Chinese and Traditional Chinese are provided. The method comprises finding for the source character a target character, for example by finding the target character in a desired data resource from the plurality of data resources which are managed by a multiple category management model with regard to data resources' priorities. The method may offer users greater flexibility in choosing the data resources most appropriate to their conversion purposes to increase the efficiency and accuracy of the conversion, and meanwhile does not have to search all the data resources before offering a conversion candidate in each operation, thereby shortening the running time of conversion.

Abstract: Various embodiments are provided for enabling audio tagging of image files. The audio messages are obtained by the system, usually by recording an audio message from a user, and then converted into a textual tag, using speech recognition technology. In some implementations semantic analysis of text component of these massages is performed. In some implementations the textual tags are then propagated to other image files associated with the user.

Abstract: Methods and systems for model-driven candidate sorting for evaluating digital evaluations are described. In one embodiment, a sorting tool selects a data set of digital evaluation data for sorting. The data set includes candidate for evaluation candidates. The sorting tool analyzes the candidate data for the respective evaluation candidate to identify digital evaluation cues and applies the digital evaluation cues to a prediction model to predict an achievement index for the respective evaluation candidate. The list of evaluation candidates is sorted according the predicted achievement indices and the sorted list is presented to the reviewer in a user interface.

Abstract: An audio encoding device includes a processor; and a memory which stores a plurality of instructions, which when executed by the processor, cause the processor to execute, calculating first phases indicating phases of a first channel signal and a second channel signal included in audio signals of a plurality of channels; and performing, on the basis of the first phases, either first predictive coding in which a third channel signal included in the audio signals of the plurality of channels is predicted using the first channel signal and the second channel signal or second predictive coding in which the second channel signal is predicted using the first channel signal.

Abstract: An apparatus for providing a control input signal for an industrial process or technical system having one or more controllable elements includes elements for generating a semantic space for a text corpus, and elements for generating a norm from one or more reference words or texts, the or each reference word or text being associated with a defined respective value on a scale, and the norm being calculated as a reference point or set of reference points in the semantic space for the or each reference word or text with its associated respective scale value. Elements for reading at least one target word included in the text corpus, elements for predicting a value of a variable associated with the target word based on the semantic space and the norm, and elements for providing the predicted value in a control input signal to the industrial process or technical system.

Abstract: Establishing a preferred mode of interaction between a user and a multimodal application, including evaluating, by a multimodal application operating on a multimodal device supporting multiple modes of interaction including a voice mode and one or more non-voice modes, user modal preference, and dynamically configuring multimodal content of the multimodal application in dependence upon the evaluation of user modal preference.

Abstract: A system for analyzing text-based information is presented. Each datum of information includes an author, a description and a timestamp. A fetcher fetches the raw information according to keywords. A parser parses the raw information to refine the results. A lexicon management module extracts lemmas from the raw information, and creates an edited lexicon containing the raw data and the lemmas for each datum. A data manager correlates lemmas in the edited lexicon and identifies clusters of lemmas that are correlated between each other. The results can be visually displayed to a user, and clusters of lemma that are less correlated than the other clusters can be visually identified. In one aspect, the user is able to excise the less correlated clusters, in order to further refine the results of the keyword search.