Abstract: A method for preparing a speech signal for encoding comprises determining whether the spectral content of an input speech signal is representative of a defined spectral characteristic (e.g., a defined characteristic slope). A frequency specific filter component of a weighting filter is controlled based on the determination of the spectral content of the speech signal or/and its location in the encoder. A core weighting filter component of the weighting filter may be maintained regardless of the spectral content of the speech signal.

Abstract: A device for operating voice-controlled systems, such as communication and/or intercommunication systems in motor vehicles, includes a plurality of microphones and at least one loudspeaker. Voice signals received by the microphones are transmitted to the at least one loudspeaker. The voice signals are subjected to a low-value frequency shift before being transmitted to the loudspeaker(s) or to the input of a voice-controlled device to thereby suppress feedback.

Abstract: In a method for performing a segmentation operation upon a synthesizing speech signal and an input speech signal, a synthesized speech signal and a speech element duration signal are generated from the synthesizing speech signal A first feature parameter is extracted from the synthesized speech signal, and a second feature parameter is extracted from the input speech signal. A dynamic programming matching operation is performed upon the second feature parameter with reference to the first feature parameter and the speech element duration signal to obtain segmentation points of the input speech signal.

Abstract: A text message is first parsed into its constituent semantic components such as header fields and body components. Then, different compression methods may be performed on each semantic component depending on the importance of the semantic component, the context of the semantic component, the characteristics of the semantic component, and whether or not the semantic component uses natural language expressions. For example, it is determined what compression method, if any, is to be performed on the semantic component. Each semantic component may be compressed individually. Since text compression takes the unique features of each semantic component into consideration rather than considering the text message as a monolithic text unit, a more intuitive text compression results.

Abstract: The present invention comprises: first periodicity providing means for emphasizing periodicity of a fixed code vector output from at least one fixed excitation code book by use of a first periodicity emphasis coefficient adaptively determined based on a predetermined rule; and second periodicity providing means for emphasizing periodicity of a fixed code vector output from at least one fixed excitation code book by use of a predetermined second periodicity emphasis coefficient.

Abstract: A system and method of recognizing speech comprises an audio receiving element and a computer server. The audio receiving element and the computer server perform the process steps of the method. The method involves training a stored set of phonemes by converting them into n-dimensional space, where n is a relatively large number. Once the stored phonemes are converted, they are transformed using single value decomposition to conform the data generally into a hypersphere. The received phonemes from the audio-receiving element are also converted into n-dimensional space and transformed using single value decomposition to conform the data into a hypersphere. The method compares the transformed received phoneme to each transformed stored phoneme by comparing a first distance from a center of the hypersphere to a point associated with the transformed received phoneme and a second distance from the center of the hypersphere to a point associated with the respective transformed stored phoneme.

Abstract: A digital audio signal to be replayed is processed in a waveform thereof. A frequency bandwidth of the audio signal is expanded through conversion of a sampling frequency, and then the audio signal is low-pass-filtered with a low-pass cut-off frequency corresponding to the converted sampling frequency. An interval of time between two waveform peaks of the audio signal is detected, and then difference data between current data of the audio signal and past data thereof is calculated. The difference data are subject to weighting depending on the interval, and then output data are produced based on both the low-pass-filtered audio signal and the weighted difference data. This processing, which can be realized by activation of software, improves audio quality when compressed audio data is replayed.

Abstract: A method, system, and program for origin device initiated caller identification are provided. In response to detecting a call extended to a destination device, extending a request from said destination device to an origin device requesting a voice utterance of the caller at said origin device. A caller identity associated with the voice utterance is identified at the destination device, such that a callee receiving the call at the destination device is informed of the caller identity before choosing whether to speak with the caller.

Abstract: An envelope generator (20), comprises: an input terminal (20a) for having a signal inputted therein; a first integrator (21) for generating intermediate state of envelopes with a first attack time and a first release time in response to changes in level of said signal inputted through said input terminal (20a) to impart said intermediate state of envelopes to said signal; a second integrator (22) for respectively modifying said intermediate state of envelopes into final state of envelopes with a second attack time and a second release time in response to changes in level of said signal imparted said intermediate state of envelopes; and an output terminal (20d) for outputting said signal with said final state of envelopes therethrough. The envelope generator (20) thus constructed can make gain signal follow rapid fluctuations in level of an audio signal, and can impart a relatively high quality for compressing and expanding level of the audio signal not to break in shape.

Abstract: A system and method for providing automatic and coordinated sharing of conversational resources, e.g., functions and arguments, between network-connected servers and devices and their corresponding applications. In one aspect, a system for providing automatic and coordinated sharing of conversational resources inlcudes a network having a first and second network device, the first and second network device each comprising a set of conversational resources, a dialog manager for managing a conversation and executing calls requesting a conversational service, and a communication stack for communicating messages over the network using conversational protocols, wherein the conversational protocols establish coordinated network communication between the dialog managers of the first and second network device to automatically share the set of conversational resources of the first and second network device, when necessary, to perform their respective requested conversational service.

Abstract: A method and system for processing, storing, retrieving and presenting information with an extendable interface for natural and artificial languages. The system includes an interpreter, a knowledge base, and an input/output module. Making use of an internal representation based on sorted-type theory, the system stores information in the knowledge base, answers queries from clients, and processes erroneous or contradictory information according to a dynamically adjustable set of rules. The system also stores language definitions in the knowledge base, enabling the system to communicate with clients in a variety of natural and artificial languages. New languages may be added to the system by presenting definitions expressed in a language already incorporated within the system.

Abstract: A method for an improved QSS (bit allocator) algorithm is disclosed. The disclosed method is capable of greatly improving determination time; thereby, improving the efficiency of converting a signal from an audio format to an MP3 format. The starting point of the QSS determination for a present frame (N) is the QSS of a previous frame (N?1). This starting point provides for improved efficiency for determining actual QSS of frame N as QSS[N?1] will be closer to QSS[N] than an arbitrary starting point. Thus, fewer iterations are required to determine QSS[N] as compared to conventional encoders. The algorithm of the present is more efficient than conventional methods in that it makes use of the fact that audio signal statistics usually do not change abruptly during the period of one audio frame to another.

Abstract: In an optical disc, a wobble signal is to be detected by a simple configuration. In the optical disc, the address information, modulated onto a sinusoidal cannier signal by adding even harmonics signals to the sinusoidal carrier signal and by changing the polarity of the harmonics signals, is formed into the wobble signal. In detecting the wobble signal from the optical disc to demodulate the address information, in a method for detecting the wobble signal, an even harmonics signal and data clocks are generated, and the even harmonics signal so generated are multiplied with the reproduced wobble signal. The resulting product signal is integrated every data clock. The sign of the digital information is verified based on the integrated value at an end edge of the data clock.

Abstract: The present invention provides a computerized method and apparatus for automatically generating from a first speech recognizer a second speech recognizer which can be adapted to a specific domain. The first speech recognizer can include a first acoustic model with a first decision network and corresponding first phonetic contexts. The first acoustic model can be used as a starting point for the adaptation process. A second acoustic model with a second decision network and corresponding second phonetic contexts for the second speech recognizer can be generated by re-estimating the first decision network and the corresponding first phonetic contexts based on domain-specific training data.

Abstract: An optical disc playback apparatus that performs appropriate waveform equalization according to a difference in characteristics of reproduced signals between land parts and groove parts in an optical disc to which high-density recording is performed using a land/groove recording method. The optical disc playback apparatus includes a reproduction device, an A/D conversion circuit for sampling a reproduced signal to be converted into a multi-bit digital signal, a filter for subjecting the multi-bit digital signal to digital equalization, a filter coefficient learning device for adaptively controlling filter coefficients for the filter corresponding to the land part and the groove part to minimize an equalization error, and a switch signal generation device for generating a signal for switching between land and groove, thereby changing the filter coefficient adaptively to the land or the groove.

Abstract: A common requirement in automatic speech recognition is to recognize a set of words for any speaker without training the system for each new speaker. A speech recognition system is provided utilizing linear discriminant based phonetic similarities with inter-phonetic unit value normalization. Linear discriminant analysis is utilized using training data with both in-class and out-class sample training utterances for generating linear discriminant vectors for each of the phonetic units. The dot product of each linear discriminant vector and the time spectral pattern vectors generated from the input speech are computed. The resultant raw similarity vectors are then normalized utilizing normalization look-up tables for providing similarity vectors which are utilized by a word matcher for word recognition.

Abstract: A method and apparatus are disclosed for transcribing speech when a number of speakers are participating. A number of different speech recognition systems, each with a different speaker model, are executed in parallel. When the identity of all of the participating speakers is known and a speaker model is available for each participant, each speech recognition system employs a different speaker model suitable for a corresponding participant. Each speech recognition system decodes the speech and generates a corresponding confidence score. The decoded output having the highest confidence score is selected for presentation to a user. When all participating speakers are not known, or when there are too many participants to implement a unique speaker model for each participant, a speaker independent speech recognition system is employed together with a speaker specific speech recognition system.

Abstract: A browser with a sound input receives a sound sequence that can be used to access a content site. The sound sequence encodes characters according to a predetermined scheme. These characters are extracted by the browser and may either directly constitute the URL of the content site, or include a site code that can be translated into the content site URL by a service system. In this latter case, the browser contacts the service system to have the site code translated into the content site URL. Once in possession of the content site URL, the browser contacts that site. Where the encoded characters represent a site code requiring translation, the URL of the translation service system is preferably included in the set of encoded characters with the encoding being such that the URL encodes to a musical tune.

Abstract: A method is provided for embedding data into an audio signal and determining data embedded into an audio signal. In the method for embedding data into an audio signal, the audio signal is based on a first set of data and includes a phase component. The method modifies at least a portion of the phase component of the audio signal to embed a second set of data into the audio signal. The modified audio signal can be made to differ with respect to the audio signal in a manner at least one of (i) substantially imperceptible and (ii) imperceptible to a listener of the first set of data depending on the extent that the phase component of the audio signal is modified. In the method for determining data embedded into an audio signal, the audio signal is based on a first set of data of an original audio signal and includes a phase component. The method determines a second set of data embedded into the audio signal based on the phase component of the audio signal.

Abstract: A pickup head actuator for carrying a plurality of near-field optical I/O elements, far-field optical I/O elements, or magnetic I/O elements. Besides performing focusing and tracking servo, the attraction due to magnetic inductor and tracking magnet enables a base to float at different heights in its normal state, suitable for different types of I/O elements.