Abstract: Various embodiments of the present invention provide systems and methods for providing a translation for a set of one or more terms or phrases related to a software application using decentralized contributions. In particular, various embodiments provide systems and methods by which multiple users of the application contribute translations for individual terms or phrases of the application instead of having one entity supply the translation of the software application. Specifically, the process of various embodiments: (1) provides a repository for storing translations of the individual terms or phrases in a target language (e.g., a language for which a translation is needed); (2) collects the translations provided by users and stores the translations in the repository; (3) identifies the preferred translation for each individual term or phrase; and (4) displays the preferred translation to the user in response to receiving a user's request for a preferred translation in the target language.

Abstract: An acquiring unit acquires pattern sentences, which are similar to one another and include fixed segments and non-fixed segments, and substitution words that are substituted for the non-fixed segments. A sentence generating unit generates target sentences by replacing the non-fixed segments with the substitution words for each of the pattern sentences. A first synthetic-sound generating unit generates a first synthetic sound, a synthetic sound of the fixed segment, and a second synthetic-sound generating unit generates a second synthetic sound, a synthetic sound of the substitution word, for each of the target sentences. A calculating unit calculates a discontinuity value of a boundary between the first synthetic sound and the second synthetic sound for each of the target sentences and a selecting unit selects the target sentence having the smallest discontinuity value. A connecting unit connects the first synthetic sound and the second synthetic sound of the target sentence selected.

Abstract: A system for combining signals includes a first microphone generating a first input signal having a first voice component and a first noise component, a second microphone generating a second input signal having a second voice component and a second noise component, a mixing circuit, and an adaptive filter. The mixing circuit applies a first gain having a value ? to the first input signal to produce a first scaled signal, applies a second gain having a value 1?? to the second input signal to produce a second scaled signal, and sums the first scaled signal and the second scaled signal to produce a summed signal. The adaptive filter computes an updated value of ? to minimize the energy of the summed signal based on the summed signal, the first input signal and the second input signal, and provides the updated value of ? to the mixing circuit.

Abstract: An encoding method and apparatus and a decoding method and apparatus are provided. The decoding method includes extracting a three-dimensional (3D) down-mix signal from an input bitstream, generating a down-mix signal with 3D effects removed therefrom by performing a 3D rendering operation on the extracted 3D down-mix signal, and generating a 3D down-mix signal with 3D effects by performing a 3D rendering operation on the generated down-mix signal. Accordingly, it is possible to efficiently encode multi-channel signals with 3D effects and to adaptively restore and reproduce audio signals with optimum sound quality according to the characteristics of an audio reproduction environment.

Abstract: A system and method for tuning a speech recognition engine to an individual microphone using a database containing acoustical models for a plurality of microphones. Microphone performance characteristics are obtained from a microphone at a speech recognition engine, the database is searched for an acoustical model that matches the characteristics, and the speech recognition engine is then modified based on the matching acoustical model.

Abstract: A method, a system, and computer readable medium comprising instructions for an open framework for speech application design are provided. The method comprises defining at least one element of a speech application, creating a design for a first step in designing the speech application using a data presentation element, storing the design in a repository using a data repository element, generating a design for a second step in designing the speech application using a data generation element, and presenting the design for the second step using the data presentation element.

Abstract: An operation control apparatus and method of controlling a plurality of operationally connected voice recognition-enabled systems, each having reciprocal control operational states corresponding to an enabled/disabled state.

Abstract: Provided herein is a character processing device that converts an input character formed of an input alphanumeric or symbol to an extended Latin character similar to the input character including: a display unit displaying as an editing character the input character with a cursor attached thereto; a conversion target distinction unit discerning whether or not the editing character is convertible to the extended Latin character; and a notification unit indicating that the editing character is convertible to the extended Latin character when the editing character is regarded as convertible.

Abstract: According to an embodiment, a translation unit translates a first sentence in a first language into a sentence in a second language using parallel translations. A next utterance table includes first identification information distinguishing between a sentence in the first language and a sentence in the second language included in the parallel translation and includes second identification information identifying a sentence previously selected as the next utterance of the sentence indicated by the first identification information. An acquiring unit acquires next utterance candidates, which are sentences indicated by the second identification information associated with the first identification information of the first sentence. If the selected next utterance candidate is the first language, the translation unit translates the selected next utterance candidate into the second language.

Abstract: In one embodiment, a method of receiving a decoded audio signal that has a transmitted pitch lag is disclosed. The method includes estimating pitch correlations of possible short pitch lags that are smaller than a minimum pitch limitation and have an approximated multiple relationship with the transmitted pitch lag, checking if one of the pitch correlations of the possible short pitch lags is large enough compared to a pitch correlation estimated with the transmitted pitch lag, and selecting a short pitch lag as a corrected pitch lag if a corresponding pitch correlation is large enough. The postprocessing is performed using the corrected pitch lag. In another embodiment, when the existence of irregular harmonics or wrong pitch lag is detected, a coded-excited linear prediction (CELP) postfilter is made more aggressive.

Abstract: Speech enhancement based on a psycho-acoustic model is disclosed that is capable of preserving the fidelity of speech while sufficiently suppressing noise including the processing artifact known as “musical noise”.

Abstract: The disclosed system and method apply stores of factual information to correct errors in digital text, for example, generated from OCR, speech and/or handwriting recognition devices, and other automatic recognition devices. A text produced by OCR, speech recognition, handwriting recognition, and others may be processed to extract discussed facts. Databases of facts are searched based on information in the text. After comparing facts asserted in the text with the factual data from the databases, suggested corrections of the text are produced.

Abstract: A logarithmic frequency spectrum within a predetermined time range is calculated from a speech signal. The logarithmic frequency spectrum has a frequency element at equal intervals along a logarithmic frequency axis. A logarithmic frequency spectrogram is calculated by connecting a plurality of logarithmic frequency spectrums. A value of the frequency element along a straight line on the logarithmic frequency spectrogram is voted onto a Hough plane. The Hough plane has a voted value in correspondence with a gradient of the straight line. The voted value above a threshold and the gradient corresponding to the voted value are extracted from the Hough plane. A fundamental frequency change is calculated using the voted value and the gradient extracted.

Abstract: A local computing device may receive an utterance from a user device. In response to receiving the utterance, the local computing device may obtain a text string transcription of the utterance, and determine a response mode for the utterance. If the response mode is a text-based mode, the local computing device may provide the text string transcription to a target device. If the response mode is a non-text-based mode, the local computing device may convert the text string transcription into one or more commands from a command set supported by the target device, and provide the one or more commands to the target device.

Abstract: This specification describes technologies relating to recognition of text in various media. In general, one aspect of the subject matter described in this specification can be embodied in methods that include receiving an input signal including data representing one or more words and passing the input signal to a text recognition system that generates a recognized text string based on the input signal. The methods may further include receiving the recognized text string from the text recognition system. The methods may further include presenting the recognized text string to a user and receiving a corrected text string based on input from the user. The methods may further include checking if an edit distance between the corrected text string and the recognized text string is below a threshold. If the edit distance is below the threshold, the corrected text string may be passed to the text recognition system for training purposes.

Abstract: Methods, systems, and apparatus are presented for decoding an audio signal that includes bandwidth extension data. An audio signal that includes core audio data and bandwidth extension data can be received in a decoder. The core audio data can be associated with a core portion of an audio signal, such as the frequency range below a cutoff frequency, and the bandwidth extension data can be associated with an extended portion of the audio signal, such as a frequency range above the cutoff frequency. The core audio data can be decoded to generate a decoded core audio signal in a time domain representation. Further, an extended portion of the audio signal can be reconstructed in accordance with extension data and decoded core audio signal. Additionally, the decoded core audio signal can be lowpass filtered and the extended portion can be highpass filtered before being combined to generate a decoded output signal.

Abstract: A method and system for performing a task in a natural language processing system. The method includes processing text to obtain a numerical encoding, where the text includes a number of entities, where at least one of the entities is tagged with the numerical encoding, and where the text is derived from an utterance. The method further includes determining that the numerical encoding identifies a task, determining a variable required to complete the task using the numerical encoding, and generating a first dialogue sentence to obtain a variable value for the variable, where the variable is associated with a second numerical encoding, where the second numerical encoding includes the numerical encoding. The method further includes transmitting the first dialogue sentence to a user device, receiving the variable value for the variable in response to the first dialogue sentence, and performing the task using the variable value.

Abstract: Systems and methods are provided to automatically determine culture-based behavioral tendencies and preferences of individuals in the context of customer service interactions. For example, systems and methods are provided to process natural language dialog input of an individual to detect linguistic features indicative of individualistic and collectivistic behavioral tendencies and predict whether such individual will be cooperative or uncooperative with automated customer service.

Abstract: Methods, systems, and apparatus, including computer program products, for translation of web documents. The original language of document text in the web document is identified. A translation header is displayed with the web document text only if the original language is different from a user's preferred language. The translation header provides the user with an option of translating the document text or portions of the document text to the user's preferred language.

Abstract: Systems, methods, and apparatuses including computer program products for universal language input are provided. In one implementation, a method is provided. The method includes using a first application or operating system component to detect user input to an application interface of a second application. The method also includes, as input is detected in the application interface, automatically using the first application to provide one or more translations of the input including, sending the input to a translation model, receiving translated input from the translation model, and sending the translated input to the application interface of the second application.