Abstract: The present document relates to audio source coding systems which make use of a harmonic transposition method for high frequency reconstruction (HFR), as well as to digital effect processors, e.g. exciters, where generation of harmonic distortion add brightness to the processed signal, and to time stretchers where a signal duration is prolonged with maintained spectral content. A system and method configured to generate a time stretched and/or frequency transposed signal from an input signal is described. The system comprises an analysis filterbank (101) configured to provide an analysis subband signal from the input signal; wherein the analysis subband signal comprises a plurality of complex valued analysis samples, each having a phase and a magnitude. Furthermore, the system comprises a subband processing unit (102) configured to determine a synthesis subband signal from the analysis subband signal using a subband transposition factor Q and a subband stretch factor 5?.

Abstract: Systems and methods are disclosed to perform relation extraction in text by applying a convolution strategy to determine a kernel between sentences; applying one or more semi-supervised strategies to the kernel to encode syntactic and semantic information to recover a relational pattern of interest; and applying a classifier to the kernel to identify the relational pattern of interest in the text in response to a query.

Abstract: An apparatus for generating a high band extension of a low band excitation signal (eLB) defined by parameters representing a CELP encoded audio signal includes the following elements: upsamplers (20) configured to upsample a low band fixed codebook vector (uFCB) and a low band adaptive codebook vector (uACB) to a predetermined sampling frequency. A frequency shift estimator (22) configured to determine a modulation frequency (?) from an estimated measure representing a fundamental frequency (F0) of the audio signal. A modulator (24) configured to modulate the upsampled low band adaptive codebook vector (uACB?) with the determined modulation frequency to form a frequency shifted adaptive codebook vector. A compression factor estimator (28) configured to estimate a compression factor. A compressor (34) configured to attenuate the frequency shifted adaptive codebook vector and the upsampled fixed codebook vector (uFCB?.) based on the estimated compression factor.

Abstract: The software application is used to rank language translations and a method to implement those ranked language translations. The language translations are for textual or vocal phrases that a user wants to convert from its original language to a preferred language. The software application will refer to a variety of translation sources in order to create a set of applicable translations for a textual or vocal phrase. The software application will then rank each applicable translation with an accuracy score, which is weighed by three factors: the commonalities between the set of translation sources, a user input process, and a linguistic mapping process. Some methods to implement the ranked language translations include using accelerometer data from a electronic communication device to operate the software application, integrating a moderator into a conversation process, using subtitles in speech bubbles for a video chat, and using location data to communicate with different contacts.

Abstract: Disclosed herein are systems, computer-implemented methods, and tangible computer-readable storage media for captioning a media presentation. The method includes receiving automatic speech recognition (ASR) output from a media presentation and a transcription of the media presentation. The method includes selecting via a processor a pair of anchor words in the media presentation based on the ASR output and transcription and generating captions by aligning the transcription with the ASR output between the selected pair of anchor words. The transcription can be human-generated. Selecting pairs of anchor words can be based on a similarity threshold between the ASR output and the transcription. In one variation, commonly used words on a stop list are ineligible as anchor words. The method includes outputting the media presentation with the generated captions. The presentation can be a recording of a live event.

Abstract: Disclosed is an apparatus for playing and producing realistic object audio. The apparatus for playing realistic object audio includes: a deformatter unit individually separating scene description (SD) compression data and object audio compression data from inputted audio files; an SD decoding unit decoding the SD compression data to restore SD information; an object audio decoding unit decoding the object audio compression data to restore object audio signals which are respective audio signals of a plurality of objects; and an object audio effect unit adding an audio effect for each object to the object audio signals according to SD information for each object corresponding to the object audio signals among the SD information to produce a realistic object audio signal corresponding to each of the object audio signals.

Abstract: An apparatus includes a monocular display with a wireless communications interface, user input device, transmitter, and controller, and may provide a video link to and control and management of a host device and other devices, such as a cell phone, computer, laptop, or media player. The apparatus may receive speech and digitize it. The apparatus may compare the digitized speech in a first language to a table of digitized speech in a second language to provide translation or, alternatively, may compare the digitized speech to a table of control commands. The control commands allow user interaction with the apparatus or other remote devices in a visual and audio manner. The control signals control a “recognized persona” or avatar stored in a memory to provide simulated human attributes to the apparatus, network or third party communication device. The avatar may be changed or upgraded according to user choice.

Abstract: A syntactic analysis and hierarchical phrase model based machine translation system and method are provided. The machine translation system includes a word alignment module, a phrase extraction module, a gender syntactic annotation module, a syntactic based noncontiguous phrase abstract module, a noncontiguous phrase based translation module and an evaluation module. The machine translation system and method performs syntactic analysis based on a common contiguous phrase based machine translation model, and extracts a syntactic based noncontiguous phrase rule set from a bilingual aligned text, so as to address and issue of noncontiguous fixed custom in context of a whole sentence, and to comply with syntactic features of a language, translation is performed based on the noncontiguous phrase rule set and the phrase aligned table, and the translation results are evaluated based on the evaluation model, thereby improving the translation result.

Abstract: An apparatus and a method for supporting many languages in a mobile communication terminal are provided. In the method, at least two installable languages are determined from a multi language image file comprising language packages of at least two languages. One of the at least two installable languages is selected and installed. The multi language image file is stored. When a language is to be changed, at least two changeable languages are determined from the multi language image file. The installed language is changed by selecting one of the at least two changeable languages. Therefore, a language to be used can be easily changed without separately downloading installation information of languages.

Abstract: Disclosed herein are systems, methods, and computer readable-media for performing an audible human verification. The method includes determining that a human verification is needed, presenting an audible challenge to a user which exploits a known issue with automatic speech recognition processes, receiving a response to the audible challenge, and verifying that a human provided the response. The known issue with automatic speech recognition processes can be recognition of a non-word, in which case the user can be asked to spell the recognized non-word. The known issue with automatic speech recognition processes can be differentiation of simultaneous input for multiple audio streams. Multiple audio streams contained in the audible challenge can be provided monaurally. Verifying that a human provided the response can include confirming the contents of one of the multiple audio streams. Audible human verification can be performed in combination with visual human verification.

Abstract: Disclosed herein are systems, methods, and computer readable-media for improving name dialer performance. The method includes receiving a speech query for a name in a directory of names, retrieving matches to the query, if the matches are uniquely spelled homophones or near-homophones, identifying information that is unique to all retrieved matches, and presenting a spoken disambiguation statement to a user that incorporates the identified unique information. Identifying information can include multiple pieces of unique information if necessary to completely disambiguate the matches. A hierarchy can establish priority of multiple pieces of unique information for use in the spoken disambiguation statement.

Abstract: In syllable or vowel or phone boundary detection during speech, an auditory spectrum may be determined for an input window of sound and one or more multi-scale features may be extracted from the auditory spectrum. Each multi-scale feature can be extracted using a separate two-dimensional spectro-temporal receptive filter. One or more feature maps corresponding to the one or more multi-scale features can be generated and an auditory gist vector can be extracted from each of the one or more feature maps. A cumulative gist vector may be obtained through augmentation of each auditory gist vector extracted from the one or more feature maps. One or more syllable or vowel or phone boundaries in the input window of sound can be detected by mapping the cumulative gist vector to one or more syllable or vowel or phone boundary characteristics using a machine learning algorithm.

Abstract: The headset comprises: a physiological sensor suitable for being coupled to the cheek or the temple of the wearer of the headset and for picking up non-acoustic voice vibration transmitted by internal bone conduction; lowpass filter means for filtering the signal as picked up; a set of microphones picking up acoustic voice vibration transmitted by air from the mouth of the wearer of the headset; highpass filter means and noise-reduction means for acting on the signals picked up by the microphones; and mixer means for combining the filtered signals to output a signal representative of the speech uttered by the wearer of the headset. The signal of the physiological sensor is also used by means for calculating the cutoff frequency of the lowpass and highpass filters and by means for calculating the probability that speech is absent.

Abstract: Semantically distinct items are extracted from a single utterance by repeatedly recognizing the same utterance using constraints provided by semantic items already recognized. User feedback for selection or correction of partially recognized utterance may be used in a hierarchical, multi-modal, or single step manner. An accuracy of recognition is preserved while the less structured and more natural single utterance recognition form is allowed to be used.

Abstract: A method for displaying a combination of languages for digital content in an electronic device is described. First digital content comprising a plurality of words in a primary language is received. Second digital content comprising a plurality of words in a secondary language is received. A combination of the primary language and the secondary language is determined. The combination of the primary language and the secondary language is displayed.

Abstract: An in-vehicle apparatus receives an image data representative of a screen image from a portable terminal with a touch panel. The apparatus extracts a text code data from the image data, and identifies a text-code display area in the screen image. The apparatus determines a command text based on a user-uttered voice command. The apparatus identifies a text-code display area as a subject operation area in the screen image of the portable terminal, based on the command text, the text code data extracted from image data, and information on the text-code display area corresponding to the text code data. An area of the screen image of the touch panel corresponding to the text-code display area is identified as the subject operation area, and a signal indicative of the subject operation area identified is transmitted to the portable terminal.

Abstract: A method and system of generating personalized annotated text of a core-text, the method and system directed for acquiring a personal data comprising data indicative of the reader's proficiency level for understanding text; and providing at least a gloss to accompany the requested text based on core-text and the reader's proficiency level.

Abstract: Systems, methods, and computer readable media providing a speech input interface. The interface can receive speech input and non-speech input from a user through a user interface. The speech input can be converted to text data and the text data can be combined with the non-speech input for presentation to a user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for recognizing speech input. In one aspect, a method includes receiving a user input and a grammar including annotations, the user input comprising audio data and the annotations providing syntax and semantics to the grammar, retrieving third-party statistical speech recognition information, the statistical speech recognition information being transmitted over a network, generating a statistical language model (SLM) based on the grammar and the statistical speech recognition information, the SLM preserving semantics of the grammar, processing the user input using the SLM to generate one or more results, comparing the one or more results to candidates provided in the grammar, identifying a particular candidate of the grammar based on the comparing, and providing the particular candidate for input to an application executed on a computing device.

Abstract: The present invention relates to a method for signal processing comprising the steps of providing a set of prototype spectral envelopes, providing a set of reference noise prototypes, wherein the reference noise prototypes are obtained from at least a sub-set of the provided set of prototype spectral envelopes, detecting a verbal utterance by at least one microphone to obtain a microphone signal, processing the microphone signal for noise reduction based on the provided reference noise prototypes to obtain an enhanced signal and encoding the enhanced signal based on the provided prototype spectral envelopes to obtain an encoded enhanced signal.