Abstract: Systems, methods, and data structures are disclosed for discovering and representing object-properties associations using textual data. The object can be a physical object or an abstract object. A system comprises a storage medium configured to store a data package containing one or more terms each comprising a word or phrase as names of properties or attributes associated with the object. The terms in the data package can collectively serve as an associative representation of the object. The data package can be used in a system for searching and classifying information based on concepts. The data package can be obtained by analyzing a plurality of text contents each containing a term defined as the object name, and other terms that are not the object name, and counting the number of occurrences of the text contents containing the object name and non-object name, or using a weighting co-efficient based on the grammatical roles of the terms, or using the frequencies of the terms in the external documents.

Abstract: A digital work of literature is vocalized using enhanced text-to-speech (TTS) controls by analyzing a digital work of literature using natural language processing to identify speaking character voice characteristics associated with context of each quote as extracted from the first work of literature; converting the character voice characteristics to audio metadata to control text-to-speech audio synthesis for each quote; transforming the audio metadata into text-to-speech engine commands, each quote being associated with audio synthesis control parameters for the TTS in the context of each the quotes in the work of literature; and inputting the commands to a text-to-speech engine to cause vocalization of the work of literature according to the words of each quote, character voice characteristics of corresponding to each quote, and context corresponding to each quote.

Abstract: A method and a device for encoding a high frequency signal, and a method and a device for decoding a high frequency signal are provided, which relate to encoding and decoding technology. The method for encoding a high frequency signal includes: determining a signal class of a high frequency signal of a current frame; smoothing and scaling time envelopes of the high frequency signal of the current frame and obtaining time envelopes of the high frequency signal of the current frame that require to be encoded, if the high frequency signal of the current frame is a non-transient signal and a high frequency signal of the previous frame is a transient signal; and quantizing and encoding the time envelopes of the high frequency signal of the current frame that require to be encoded, and frequency information and signal class information of the high frequency signal of the current frame.

Abstract: A device may receive an audio sample, and may separate the audio sample into multiple sub-band signals in multiple frequency bands. The device may modify an upper boundary and a lower boundary of at least one of the frequency bands to form modified frequency bands. The device may modify the sub-band signals to form banded signals associated with the modified frequency bands. The device may smooth the banded signals to form smoothed signal values. The device may identify peak values included in the smoothed signal values, and may generate an audio fingerprint for the audio sample based on the smoothed signal values and the peak values.

Abstract: Provided are a method and apparatus for encoding and decoding an audio signal. According to the present application, a signal of a high frequency band above a preset frequency band is adaptively encoded or decoded in the time domain or in the frequency domain by using a signal of a low frequency band below the preset frequency band. As such, the sound quality of a high frequency signal is not deteriorate even when an audio signal is encoded or decoded by using a small number of bits and thus coding efficiency may be maximized.

Abstract: An electronic apparatus and a method of controlling the electronic apparatus are provided. The method includes: receiving a voice command; and if the voice command is a first voice start command, changing a mode of the electronic apparatus to a first voice task mode in which the electronic apparatus is controlled according to further voice input, and if the voice command is a second voice start command, changing the mode of the electronic apparatus to a second voice task mode in which the electronic apparatus is controlled according to the further voice input received via an external apparatus which operates with the electronic apparatus. Therefore, providing efficiency and flexibility in controlling the electronic apparatus by using a microphone of the electronic apparatus or a microphone of the external apparatus.

Abstract: The present disclosure relates to a mobile terminal and a voice recognition method thereof. The voice recognition method may include receiving a user's voice; providing the received voice to a first voice recognition engine provided in the server and a second voice recognition engine provided in the mobile terminal; acquiring first voice recognition data as a result of recognizing the received voice by the first voice recognition engine; acquiring second voice recognition data as a result of recognizing the received voice by the second voice recognition engine; estimating a function corresponding to the user's intention based on at least one of the first and the second voice recognition data; calculating a similarity between the first and the second voice recognition data when personal information is required for the estimated function; and selecting either one of the first and the second voice recognition data based on the calculated similarity.

Abstract: A system and computer-implemented method that includes analyzing content of textual information, automatically deriving a plurality of questions from the analyzed content, presenting the plurality of questions derived, to a user via a user interface, determining, via the user, a validity of the plurality of questions, and generating at least one valid question as determined by a user along with a corresponding answer to at least one valid question. The method further includes generating a summary of the textual information using the answers generated. The method further includes the ability of generate grammars for parsing natural language phrases.

Abstract: In response to a signal failing to exceed an estimated level of noise by more than a predetermined amount for more than a predetermined continuous duration, the estimated level of noise is adjusted according to a first time constant in response to the signal rising and a second time constant in response to the signal falling, so that the estimated level of noise falls more quickly than it rises. In response to the signal exceeding the estimated level of noise by more than the predetermined amount for more than the predetermined continuous duration, a speed of adjusting the estimated level of noise is accelerated.

Abstract: A voice recognition apparatus includes a command recognizer and a data recognizer. The command recognizer recognizes a command portion of a voice input and outputs a command based on a voice recognition result of the voice input. The data recognizer recognizes a data portion of a voice inputs and outputs a data based on a voice recognition result of the voice input. The data recognizer further includes a plurality of data-category recognizers respectively using a data-category dictionary for recognizing the data portion of the voice input and outputting a data result. A voice recognition result selection unit of the voice recognition apparatus selects one of the data results from the data-category recognizers based on the command recognized by the command recognizer.

Abstract: Respective word frequencies may be determined from a corpus of utterance-to-text-string mappings that contain associations between audio utterances and a respective text string transcription of each audio utterance. Respective compressed word frequencies may be obtained based on the respective word frequencies such that the distribution of the respective compressed word frequencies has a lower variance than the distribution of the respective word frequencies. Sample utterance-to-text-string mappings may be selected from the corpus of utterance-to-text-string mappings based on the compressed word frequencies. An automatic speech recognition (ASR) system may be trained with the sample utterance-to-text-string mappings.

Abstract: A system and method for entering text from a user includes a programmed processor that receives inputs from the user and disambiguates the inputs to present word choices corresponding to the text. In one embodiment, inputs are received in two or more modalities and are analyzed to present the word choices. In another embodiment, a keyboard is divided into zones each of which represents two more input characters. A sequence of zones selected by the user is analyzed to present word choices corresponding to the zone selected.

Abstract: An electronic apparatus and a method of controlling the electronic apparatus are provided. The method includes: receiving a voice command; and if the voice command is a first voice start command, changing a mode of the electronic apparatus to a first voice task mode in which the electronic apparatus is controlled according to further voice input, and if the voice command is a second voice start command, changing the mode of the electronic apparatus to a second voice task mode in which the electronic apparatus is controlled according to the further voice input received via an external apparatus which operates with the electronic apparatus. Therefore, providing efficiency and flexibility in controlling the electronic apparatus by using a microphone of the electronic apparatus or a microphone of the external apparatus.

Abstract: Methods, apparatus and computer program products implement embodiments of the present invention that include receiving a time domain voice signal, and extracting a single pitch cycle from the received signal. The extracted single pitch cycle is transformed to a frequency domain, and the misclassified roots of the frequency domain are identified and corrected. Using the corrected roots, an indication of a maximum phase of the frequency domain is generated.

Abstract: A method for producing speech recognition results on a device includes receiving first speech recognition results, obtaining a language model, wherein the language model represents information stored on the device, and using the first speech recognition results and the language model to generate second speech recognition results.

Abstract: A content alignment service may generate content synchronization information to facilitate the synchronous presentation of audio content and textual content. In some embodiments, a region of the textual content whose correspondence to the audio content is uncertain may be analyzed to determine whether the region of textual content corresponds to one or more words that are audibly presented in the audio content, or whether the region of textual content is a mismatch with respect to the audio content. In some embodiments, words in the textual content that correspond to words in the audio content are synchronously presented, while mismatched words in the textual content may be skipped to maintain synchronous presentation. Accordingly, in one example application, an audiobook is synchronized with an electronic book, so that as the electronic book is displayed, corresponding words of the audiobook are audibly presented.

Abstract: A method for configuring a voice authentication system employing at least one authentication engine comprises utilizing the at least one authentication engine to systematically compare a plurality of impostor voice sample against a voice sample of a legitimate person to derive respective authentication scores. The resultant authentication scores are analyzed to determine a measure of confidence for the voice authentication system.

Abstract: The text of an incoming message may be scanned for language that matches in whole or in part the name of a stored contact and/or information indicative of a stored location. Each identified instance of language may be converted into a link. When actuated, the link may initiate an action relating to the information to which the identified language was matched, such as to call or send an email to a matching contact or to show a map of or navigating instructions to the location indicated by matching location language. When the identified language matches multiple contacts or is indicative of multiple locations, actuation of the link may result in the display of a menu of all of the matches so that the desired one can be selected. A list of possible actions may also be provided.

Abstract: A method of removing a noise signal from an input signal, the method including receiving a pure noise signal and an input signal including the noise signal; determining whether the pure noise signal is a static noise signal or a non-static noise signal; and removing the noise signal from the input signal according to whether the noise signal is determined to be the static noise signal or the non-static noise signal.

Abstract: A probability model represented as the product of the probability distribution of a mismatch vector g (or clean speech x) with an observed value y as a factor and the probability distribution of a mismatch vector g (or clean speech x) with a confidence index ? for each band as a factor, executes MMSE estimation on the probability model, and estimates a clean speech estimated value x^. As a result, each band influences the result of MMSE estimation, with a degree of contribution in accordance with the level of its confidence. Further, the higher the S/N ratio of observation speech, the more the output value becomes shifted to the observed value. As a result, the output of a front-end is optimized.