Abstract: Signals of different channels are combined into one mono signal. A set of adaptive filters, preferably one for each channel, is derived in a respective filter adaptation unit. When an adaptive filter is applied to the mono signal it reconstructs the signal of the respective channel under a perceptual constraint. The perceptual constraint is a gain and/or shape constraint. The gain constraint allows the preservation of the relative energy between the channels while the shape constraint allows more stability by avoiding unnecessary filtering of spectral nulls. The transmitted parameters are the mono signal, in encoded form, and the parameters of the adaptive filters, preferably also encoded. The receiver reconstructs the signal of the different channels by applying the adaptive filters and possibly some additional post-processing.

Abstract: A spectral representation of an audio signal having consecutive audio frames can be derived more efficiently, when a common time warp is estimated for any two neighboring frames, such that a following block transform can additionally use the warp information. Thus, window functions required for successful application of an overlap and add procedure during reconstruction can be derived and applied, the window functions already anticipating the re-sampling of the signal due to the time warping. Therefore, the increased efficiency of block-based transform coding of time-warped signals can be used without introducing audible discontinuities.

Abstract: The present invention provides a chat information system having a voice recognition device for recognizing voices, a voice synthesizer, a humanoid robot, a microphone for receiving the voices and a speaker for pronouncing synthesized voices. The system comprises a headline sensor capturing news from the Internet, a news database for storing the captured news, and a conversation database including at least a general conversation database storing a set of inquiries and responses. The system also includes a chat engine configured to extract one or more keywords from a user's speech that has been recognized by the voice recognition device, to search at least one of the news database and the conversation database with the extracted keywords and to output via the speaker the contents that have been hit by the search.

Abstract: An approach to scoring acoustically-based events, such as hypothesized instances of keywords, in a speech processing system make use of scores of individual components of the event. Data characterizing an instance of an event are first accepted. This data includes a score for the event. The event is associated with a number of component events from a set of component events, such as a set of phonemes. Probability models are also accepted for component scores associated with each of the set of component events in each of two of more possible classes of the event, such as a class of true occurrences of the event and a class of false detections of the event. The event is then scored. This scoring includes computing a probability of one of the two or more possible classes for the event using the accepted probability models.

Abstract: Coding an audio signal wherein values of first parameters, which represent aspects of the audio signal at a first instant are calculated to obtain first calculated values and values of second parameters, which represent the aspects of the audio signal at a second, later, instant, are calculated to obtain second calculated values, wherein the number of the first parameters and the number of the second parameters differ. The values of the subset of the second parameters are coded based on a difference of this subset and a subset of the first calculated value associated with substantially a same particular portion of the frequency range. Thus the differentially coded values of the second parameters are obtained by coding the difference of the values of second parameters and first parameters which are associated with substantially the same frequency sub-range.

Abstract: A system receives a spoken utterance, identifies at least one keyword within the spoken utterance, and identifies a function using the identified at least one keyword. The system further performs the identified function on at least a portion of the spoken utterance to create a voice file.

Abstract: A speech recognition apparatus and method of this invention manage previously input frequencies of occurrence for respective geographical names to be recognized (202), update the probability of occurrence of the geographical name to be recognized of interest on the basis of the frequency of occurrence of that geographical name, and those of geographical names to be recognized located within a predetermined region including the position of the geographical name of interest using a table (114) that describes correspondence between the geographical names to be recognized and their positions, and perform this update process for respective geographical names to be recognized (203).

Abstract: A coded signal conveys encoded audio information and metadata that may be used to control the loudness and dynamic range of the audio information during its playback. If the values for these metadata parameters are set incorrectly, annoying fluctuations in loudness during playback can result. The present invention overcomes this problem by detecting incorrect metadata parameter values in the signal and replacing the incorrect values with corrected values.

Abstract: Operations for weighted and non-weighted multi-tape automata are described for use in natural language processing tasks such as morphological analysis, disambiguation, and entity extraction.

Abstract: In one alternative, an audio signal is analyzed using multiple psychoacoustic criteria to identify a region of the signal in which time scaling and/or pitch shifting processing would be inaudible or minimally audible, and the signal is time scaled and/or pitch shifted within that region. In another alternative, the signal is divided into auditory events, and the signal is time scaled and/or pitch shifted within an auditory event. In a further alternative, the signal is divided into auditory events, and the auditory events are analyzed using a psychoacoustic criterion to identify those auditory events in which the time scaling and/or pitch shifting procession of the signal would be inaudible or minimally audible. Further alternatives provide for multiple channels of audio.

Abstract: Embodiments of a system, method and computer program product for verifying an identity claimed by a claimant using voiced to unvoiced classifiers are described. In accordance with one embodiment, a speech sample from a claimant claiming an identity may be captured. From the speech sample, a ratio of unvoiced frames to a total number of frames in the speech sample may be calculated. An equal error rate value corresponding to the speech sample can then be determined based on the calculated ratio. The determined equal error rate value corresponding to the speech sample may be compared to an equal error rate value associated with the claimed identity in order to select a decision threshold. A match score may be also be generated based on a comparison of the speech sample to a voice sample associated with the claimed identity. A decision whether to accept the identity claim of the claimant can then be made based on a comparison of the match score to the decision threshold.

Abstract: A sentence is accessed and at least one query is generated based on the sentence. At least one query can be compared to text within a collection of documents, for example using a web search engine. Collocation errors in the sentence can be detected and/or corrected based on the comparison of the at least one query and the text within the collection of documents.

Abstract: A command processing apparatus includes an LCD. A manual trick action instruction to a dog object displayed on the LCD is accepted through a touch panel. The dog object performs a trick according to the accepted trick action instruction. An arbitrary voice command is fetched by a microphone in relation to the trick action. A voice command being coincident with the fetched voice command is retrieved from among the registered voices in a RAM through a voice verification process by a CPU core. When the verification process fails, a verification result indicates “?1”. The fetched voice command is assigned to the current trick. On the other hand, if the voice command found by the verification process is the voice command that is assigned to the current trick, a degree of relation corresponding to the trick is incremented. The dog object performs a different action depending on the degree of relation.

Abstract: A harmonic structure acoustic signal detection device not depending on the level fluctuation of the input signal including: an FFT unit which performs FFT on an input signal and calculates a power spectrum component for each frame; a harmonic structure extraction unit which leaves only a harmonic structure from the power spectrum component; a voiced feature evaluation unit which evaluates correlation between the frames of harmonic structures extracted by the harmonic structure extraction unit, thereby evaluates whether or not the segment is a vowel segment, and extracts the voiced segment; and a speech segment determination unit which determines a speech segment according to the continuity and durability of the output of the voiced feature evaluation unit.

Abstract: The present invention utilizes a discriminative density model selection method to provide an optimized density model subset employable in constructing a classifier. By allowing multiple alternative density models to be considered for each class in a multi-class classification system and then developing an optimal configuration comprised of a single density model for each class, the classifier can be tuned to exhibit a desired characteristic such as, for example, high classification accuracy, low cost, and/or a balance of both. In one instance of the present invention, error graph, junction tree, and min-sum propagation algorithms are utilized to obtain an optimization from discriminatively selected density models.

Abstract: A method, a system, and an apparatus biasing a speech recognizer based on prompt context. The present invention is capable of analyzing the words used in the prompt given to the user. Then, a set of words the user is likely to say in response to the prompt is determined. The word set may be determined using a technology used by the speech recognition system, such as n-grams, grammars, or both. The speech recognition system boosts the probabilities of the analyzed words in the word set by a preconfigured amount. The preconfigured amount is selected based on collected data.

Abstract: The invention relates to a network element (1) and a method for enhancing the quality of digitised analogue signals transmitted in parameterised coded form via a digital network. In order to enable an enhancement of the quality of the digitised analogue signals on network side, the network element comprises means (20, 21) for extracting signals from and insert signals into the network, first processing means (24) for processing the extracted parameters in the parameter domain with functions suitable to enhance the quality of the digitised analogue signals and second processing means (26) for processing the extracted parameters in the linear domain with functions suitable to enhance the quality of the digitised analogue signals. Moreover included analysing and selecting means (23, 27) determine the expected enhancement of quality in the different processing domains and cause a corresponding insertion of processed signals back into the network. The proposed method comprises corresponding steps.

Abstract: Provided are a flexible bit rate code vector generation method and a wideband vocoder employing the same. This invention implements a flexible bit rate by getting three code vectors which are composed of 24, 16, and 8 pulses, at a time in a search process, through improvement of an algebraic codebook search process in a wideband AMR-WB vocoder. The method includes the steps of: performing a preprocess, wherein the preprocess divides a sub-frame by tracks and decides a pulse position having a maximum value in each track; among a plurality of pulses to be searched, fixing a same number of pulses as the tracks to the position with the maximum value of each track sequentially, and searching optimal positions having a minimum error with a target signal by combining two pulses in two consecutive tracks for the remaining pulses; and creating a code vector with flexible bit rate.

Abstract: An interface system including a manipulandum adapted to be moveable according to a manual gesture imparted by the user; a sensor adapted to detect a characteristic of the manual gesture imparted to the manipulandum and to generate a sensor signal representing the detected characteristic of the manual gesture; a microphone adapted to detect a characteristic of an utterance spoken by the user and to generate an audio signal representing the detected characteristic of the spoken utterance; and a control system adapted receive the generated sensor and audio signals and to transmit a command signal to an electronic device via a communication link, the command signal being based on the generated sensor and audio signals and the time synchronization between them.

Abstract: A control unit of an electronic-book read-aloud device reads book data and electronic-book data from an electronic bookmark and stores the read data in a storage unit. Further, the control unit sets a read-aloud-start position based on the electronic-bookmark data, reads the book data after the read-aloud-start position from the storage unit, and transmits the book data to a speech-output unit. The speech-output unit converts the book data into a speech signal and transmits the speech signal to a speaker through an amplifier. If the read-aloud processing is stopped, the control unit writes read-aloud-end-position data and read-aloud-date data into the electronic bookmark of the electronic book.