Abstract: Various embodiments of systems, methods, and computer programs are disclosed for providing real-time resources to participants in an audio conference session. One embodiment is a method for providing real-time resources to participants in an audio conference session via a communication network. One such method comprises: a conferencing system establishing an audio conference session between a plurality of computing devices via a communication network, each computing device generating a corresponding audio stream comprising a speech signal; and in real-time during the audio conference session, a server: receiving and processing the audio streams to determine the speech signals; extracting words from the speech signals; analyzing the extracted words to determine a relevant keyword being discussed in the audio conference session; identifying a resource related to the relevant keyword; and providing the resource to one or more of the computing devices.

Abstract: A system for automated adaptation and improvement of speaker authentication in a voice biometric system environment, comprising a speech sample collector, a target selector, a voice analyzer, a voice data modifier, and a call flow creator. The speech sample collector retrieves speech samples from a database of enrolled participants in a speaker authentication system. The target selector selects target users that will be used to test the speaker authentication system. The voice analyzer extracts a speech component data set from each of the speech samples. The call flow creator creates a plurality of call flows for testing the speaker authentication system, each call flow being either an impostor call flow or a legitimate call flow. The call flows created by the call flow creator are used to test the speaker authentication system.

Abstract: Methods and devices for efficient encoding/decoding of a time segment of an audio signal. The methods comprise deriving an indicator, z, of the position in a frequency scale of a residual vector associated with the time segment of the audio signal, and deriving a measure, ?, related to the amount of structure of the residual vector. The methods further comprise determining whether a predefined criterion involving the measure ?, the indicator z and a predefined threshold ?, is fulfilled, which corresponds to estimating whether a change of sign of at least some of the non-zero coefficients of the residual vector would be audible after reconstruction of the audio signal time segment. The respective amplitude of the coefficients of the residual vector is encoded, and the signs of the coefficients of the residual vector are encoded only when it is determined that the criterion is fulfilled, and thus that a change of sign would be audible.

Abstract: Disclosed are systems, methods and computer-readable media for enabling speech processing in a user interface of a device. The method includes receiving an indication of a field and a user interface of a device, the indication also signaling that speech will follow, receiving the speech from the user at the device, the speech being associated with the field, transmitting the speech as a request to public, common network node that receives and processes speech, processing the transmitted speech and returning text associated with the speech to the device and inserting the text into the field. Upon a second indication from the user, the system processes the text in the field as programmed by the user interface. The present disclosure provides a speech mash up application for a user interface of a mobile or desktop device that does not require expensive speech processing technologies.

Abstract: A method and related system, computer program product and device for interactively tracking oral reading of text from a document includes recording audio for a sentence read by a user and determining when the user has reached the last word of the sentence. The method also includes providing visual feedback to the user reading on a sentence by sentence level to indicate a current location in the passage.

Abstract: An audio packet error concealment system includes an encoding unit for encoding an audio signal consisting of a plurality of frames, and an auxiliary information encoding unit for estimating and encoding auxiliary information about a temporal change of power of the audio signal. The auxiliary information is used in packet loss concealment in decoding of the audio signal. The auxiliary information about the temporal change of power may contain a parameter that functionally approximates a plurality of powers of subframes shorter than one frame, or may contain information about a vector obtained by vector quantization of a plurality of powers of subframes shorter than one frame.

Abstract: An automatic speech recognition (ASR) system includes a speech-responsive application and a recognition engine. The ASR system generates user prompts to elicit certain spoken inputs, and the speech-responsive application performs operations when the spoken inputs are recognized. The recognition engine compares sounds within an input audio signal with phones within an acoustic model, to identify candidate matching phones. A recognition confidence score is calculated for each candidate matching phone, and the confidence scores are used to help identify one or more likely sequences of matching phones that appear to match a word within the grammar of the speech-responsive application. The per-phone confidence scores are evaluated against predefined confidence score criteria (for example, identifying scores below a ‘low confidence’ threshold) and the results of the evaluation are used to influence subsequent selection of user prompts.

Abstract: Provided are a method, system, and computer program product for handling locale and language in a cloud management system, in which a first composite values list of applicable locales and matching languages combinations is generated from at least one language installed on a service management system and at least one locale supported by said service management system. A second composite values list of applicable locales and matching languages combinations is generated as a fall back list based on at least one base language of said service management system and at least one matching locale formed from said at least one base language, if said first composite values list of applicable locales and matching languages is empty. A resulting composite values list of valid locales and languages combinations is provided for further processing.

Abstract: The technology of the present application provides a method and apparatus to allow for dynamically updating a language model across a large number of similarly situated users. The system identifies individual changes to user profiles and evaluates the change for a broader application, such as, a dialect correction for a speech recognition engine, as administrator for the system identifies similarly situated user profiles and downloads the profile change to effect a dynamic change to the language model of similarly situated users.

Abstract: An audio signal processing apparatus that processes a bit stream generated by coding an audio signal on a frame-by-frame basis, the bit stream including, for each frame, coded data representing the audio signal, additional data and attribute information, the audio signal processing apparatus including a decoding unit configured to decode the coded data to generate a decoded signal, a processing unit configured to process the decoded signal, a detection unit configured to detect whether or not there has been a change in the attribute information, and a storage unit, wherein the processing unit is configured to, when the change is not detected, process the decoded signal by using at least two pieces of additional data stored, and when the change is detected, process the decoded signal by using only either additional data before detection of the change or additional data after detection of the change.

Abstract: Various techniques for autocorrecting virtual keyboard input for various languages (e.g., Japanese, Chinese) are disclosed. In one aspect, a system or process receives a sequence of keyboard events representing keystrokes on a virtual keyboard. A hierarchical data structure is traversed according to the sequence of keyboard events to determine candidate words for the sequence of keyboard events. A word lattice is constructed using a language model, including deriving weights or paths in the word lattice based on candidate word statistics and data from a keyboard error model. The word lattice is searched to determine one or more candidate sentences comprising candidate words based on the path weights. Paths through the word lattice can be pruned (e.g., discarded) to reduce the size and search time of the word lattice.

Abstract: The present invention addresses the deficiencies in the prior art by providing an improved dialog for disambiguating a user utterance containing more than one intent. The invention comprises methods, computer-readable media, and systems for engaging in a dialog. The method embodiment of the invention relates to a method of disambiguating a user utterance containing at least two user intents. The method comprises establishing a confidence threshold for spoken language understanding to encourage that multiple intents are returned, determining whether a received utterance comprises a first intent and a second intent and, if the received utterance contains the first intent and the second intent, disambiguating the first intent and the second intent by presenting a disambiguation sub-dialog wherein the user is offered a choice of which intent to process first, wherein the user is first presented with the intent of the first or second intents having the lowest confidence score.

Abstract: An audio encoder implements multi-channel coding decision, band truncation, multi-channel rematrixing, and header reduction techniques to improve quality and coding efficiency. In the multi-channel coding decision technique, the audio encoder dynamically selects between joint and independent coding of a multi-channel audio signal via an open-loop decision based upon (a) energy separation between the coding channels, and (b) the disparity between excitation patterns of the separate input channels. In the band truncation technique, the audio encoder performs open-loop band truncation at a cut-off frequency based on a target perceptual quality measure. In multi-channel rematrixing technique, the audio encoder suppresses certain coefficients of a difference channel by scaling according to a scale factor, which is based on current average levels of perceptual quality, current rate control buffer fullness, coding mode, and the amount of channel separation in the source.

Abstract: The present invention proposes a new method and a new apparatus for enhancement of audio source coding systems utilizing high frequency reconstruction (HFR). It utilizes a detection mechanism on the encoder side to assess what parts of the spectrum will not be correctly reproduced by the HFR method in the decoder. Information on this is efficiently coded and sent to the decoder, where it is combined with the output of the HFR unit.

Abstract: Remote controllers and systems thereof are disclosed. The remote controller remotely operates a receiving host, in which the receiving host provides voice input and speech recognition functions. The remote controller comprises a first input unit and a second input unit for generating a voice input request and a speech recognition request. The generated voice input and speech recognition requests are then sent to the receiving host, thereby forcing the receiving host to perform the voice input and speech recognition functions.

Abstract: The disclosure provides a method and an apparatus for detecting a voice activity in an input audio signal composed of frames. A noise characteristic of the input signal is determined based on a received frame of the input audio signal. A voice activity detection (VAD) parameter is derived based on the noise characteristic of the input audio signal using an adaptive function. The derived VAD parameter is compared with a threshold value to provide a voice activity detection decision. The input audio signal is processed according to the voice activity detection decision.

Abstract: A method, computer system, and computer program product for translating information. The computer system receives the information for a translation. The computer system identifies portions of the information based on a set of rules for security for the information in response to receiving the information. The computer system sends the portions of the information to a plurality of translation systems. In response to receiving translation results from the plurality of translation systems for respective portions of the information, the computer system combines the translation results for the respective portions to form a consolidated translation of the information.

Abstract: Techniques are described herein for automatic generation of a title or summary from a long body of text. A grammatical tree representing one or more sentences of the long body of text is generated. One or more nodes from the grammatical tree are selected to be removed. According to one embodiment, a particular node is selected to be removed based on its position in the grammatical tree and its node-type, where the node type represents a grammatical element of the sentence. Once the particular node is selected, a branch of the tree is cut at the node. After branch has been cut, one or more sub-sentences are generated from the remaining nodes in the grammatical tree. The one or more sub-sentences may be returned as a title or summary.

Abstract: An apparatus and method for tracking dialogue and other sound signals in film, television or other systems with multiple channel sound is described. One or more audio channels which is expected to carry the speech of persons appearing in the program or other particular types of sounds is inspected to determine if that channel's audio includes particular sounds such as MUEVs, including phonemes corresponding to human speech patterns. If an improper number of particular sounds such as phonemes are found in the channel(s) an action such as a report, an alarm, a correction, or other action is taken. The inspection of the audio channel(s) may be made in conjunction with the appearance of corresponding images associated with the sound, such as visemes in the video signal, to improve the determination of types of sounds such as phonemes.

Abstract: Methods and systems for translating a text into multiple languages performed by at least one software component executed by at least one processor, comprise: maintaining a translation repository having a plurality of entries associating different types of content with user-specified languages; monitoring the text received by a program to identify one or more types of content and a source language of the text; retrieving the user-specified languages from the translation repository associated with the identified types of content; and for each of the identified types of content, translating the content thereof from the source language to the corresponding user-specified language when the source language is different from the corresponding user-specified language.