Abstract: Disclosed is a system and method for training a spoken dialog service component from website data. Spoken dialog service components typically include an automatic speech recognition module, a language understanding module, a dialog management module, a language generation module and a text-to-speech module. The method includes converting data from a structured database associated with a website to a structured text data set and a structured task knowledge base, extracting linguistic items from the structured database, and training a spoken dialog service component using at least one of the structured text data, the structured task knowledge base, or the linguistic items. The system includes modules configured to implement the method.

Abstract: A method, medium, and apparatus encoding/decoding audio data in which audio data is hierarchically encoded, and at least one extension data of the audio data is encoded using at least one encoding method, and decoding is performed in the same manner, thereby ensuring fine grain scalability (FGS) and unlimited extendibility of the audio data.

Abstract: Provided is an audio encoding and decoding apparatus and method for improving a compression ratio while maintaining sound quality when sinusoidal waves of an audio signal are connected and encoded. The audio encoding method includes connecting sinusoidal waves of an input audio signal, converting a frequency of each of the connected sinusoidal waves to a psychoacoustic frequency, performing a first encoding operation for encoding the psychoacoustic frequency, performing a second encoding operation for encoding an amplitude of each of the connected sinusoidal waves, and outputting an encoded audio signal comprising the encoding result of the first encoding operation and the encoding result of the second encoding operation.

Abstract: A receiver in an audio coding system receives a signal conveying frequency subband signals representing an audio signal. The subband signals are examined to assess one or more characteristics of the audio signal including temporal shape. Spectral components are synthesized having the one or more assessed characteristics, integrated with the subband signals and passed through a synthesis filterbank to generate an output signal.

Abstract: In one embodiment, sample information and frame length information are obtained from the audio signal. The sample information indicates a total number of audio data samples for each channel in the audio signal, and the frame length information indicates a number of samples in a frame of each channel. An optimum prediction order is determined for each block based on a maximum permitted prediction order and a length of the block, where a prediction order is the number of linear prediction coefficients. The optimum prediction order is selected as a minimum one of the global prediction order and the local prediction order. The global prediction order is determined based on the maximum permitted prediction order, and the local prediction order is determined based on the length of the block.

Abstract: A large-scale attendance, productivity, activity and availability biometric control method using the telephone network, for individual client users with speaker verification technology based on limited enrolling data and short verification sentences.

Abstract: A driving directions system loads into memory a limited subset of prerecorded, spoken utterances of geographic names from a mass media storage. The subset of spoken utterances may be limited, for example, to the geographic names within a predetermined radius (e.g., a few miles) of the driver's present location. The present location of the driver may be manually entered into the driving directions system by the driver, or automatically determined using a global positioning system (“GPS”) receiver. As the vehicle moves from its present location, the driving directions system loads into memory new names from the mass media storage and overwrites, if necessary, those which are now geographically out of range. Based on the current location of the driving, the driving directions system can audibly output geographic names from the run-time memory.

Abstract: An audio decoder which reproduces original signals from a bit stream including (i) a downmix signal of the original signals, and (ii) supplementary information indicating a gain ratio D and phase difference ? between the original signals. The audio decoder includes: a decoding unit extracting the downmix signal; a transformation unit transforming the extracted downmix signal into a frequency domain signal; a phase rotator determination unit determining two phase rotators having, as the phase rotation angles, angles ? and ? respectively obtained by dividing a contained angle by a diagonal of a parallelogram; a separation unit separating the frequency domain signal into two separation signals respectively indicating angles ? and ? as phase differences between the signals and the decoded downmix signal; and an inverse transformation unit inversely transforming the respective two separation signals into time domain signals so as to reproduce the two audio signals.

Abstract: Echo residue is detected using speech application data. The echo residue is detected using a method that includes correlating audio data from an input channel with audio data from an output channel to obtain a correlation result. A determined value of the correlation result is compared with a predetermined threshold. The audio data for the input channel is categorized in a first category when the determined value of the correlation result is greater than the predetermined threshold. The audio data for the input channel is categorized in a second category when the determined value of the correlation result is less than the predetermined threshold. The first category includes audio data that is determined to include an acceptable level of residual echo. The second category includes audio data that is determined to include an unacceptable level of residual echo.

Abstract: Exemplary embodiments are directed to a device and a computer-readable storage medium for creating and editing documents or messages by dynamically loading the required data on the computing device as the documents or messages are being created or edited. These exemplary embodiments have relevance for creating or editing documents or messages in non-English languages using a computing device that is pre-configured to create English documents or messages, but not non-English documents or messages. Further, these embodiments allow a user to create and edit documents and messages on a computing device that may not have been configured a priori or have limited storage capability to support the entire data set required for creating the documents or messages in a specific language. The computing device is required to communicate with a data storage device to dynamically load the required data from therein.

Abstract: A server for providing language and literacy tutoring information to a plurality of user devices connected to a communications network, comprising a network adapter for connection to the network; a content database for providing learning content to devices via the network adaptor and the network; a plurality of speech recognition models stored in the server; a processor for processing speech data and session control data generated by a user and sent to the server by the network, the processor evaluating which of the speech recognition models provides most accurate speech recognition results; and a performance evaluator for evaluating speech produced by the user using the speech recognition model that produces the most accurate results. A system, including user devices. A method for operating the system, and a program storage medium having computer code thereon for implementing the method and system.

Abstract: Natural language interface to a back-end application, incorporating synonyms, suggestions, and proposals. Roughly described, synonyms are automatically added to user input to enhance the natural language interpretation, whereas suggestions and proposals are offered to the user in an interaction, usually after an interpretation of prior user input. Suggestions and synonyms can be learned from user input, whereas proposals are programmed by a third party. The selection of synonyms, suggestions, and proposals for use with particular user input can be user input context-based so that further user input can maintain context by explicitly indicating that the same context is intended, and rewards-based reinforcement can be used to better focus suggestions and proposals on the characteristics of the particular user.

Abstract: Methods and systems for providing a network-accessible text-to-speech synthesis service are provided. The service accepts content as input. After extracting textual content from the input content, the service transforms the content into a format suitable for high-quality speech synthesis. Additionally, the service produces audible advertisements, which are combined with the synthesized speech. The audible advertisements themselves can be generated from textual advertisement content.

Abstract: A keypad is used to enter complex characters using a phonetic input method editor (IME). The user may enter complex characters by combining consonants, vowels, mid-vowels and tones by selecting keys on a the keypad instead of using a full size keyboard. Instead of a one-to-one mapping between the symbols and keys on a full size keyboard, multiple symbols are assigned to single keys on the keypad. For example, on a keypad having ten keys an average of four phonetic symbols are mapped to each of the ten keys on the keypad. The phonetic symbols are applied to the keypad in layers. For example, the symbols may be may be mapped to a consonant layer; a middle vowels+vowels layer; a vowels layer and a tone layer. Phonetic symbols with similar readings may also be mapped to the same key.

Abstract: A receiver in an audio coding system receives a signal conveying frequency subband signals representing an audio signal. The subband signals are examined to assess one or more characteristics of the audio signal including temporal shape. Spectral components are synthesized having the one or more assessed characteristics, integrated with the subband signals and passed through a synthesis filterbank to generate an output signal.

Abstract: A source sampling rate is associated with first or second groups of sampling rates. A playback rate is determined by: (a) selecting the source sampling rate if the source sampling rate is supported by a playback environment; (b) otherwise if there is a highest first rate from the first or second groups of playback sampling rates which is supported by the playback environment and is lower than the source sampling rate, selecting the first rate; (c) otherwise if there is a slowest second rate from the group that the source sampling rate is associated with that is supported by the playback environment and is higher than the source sampling rate, selecting the second rate; (d) otherwise selecting the slowest sampling rate supported by the playback environment from the group that the source sampling rate is not associated with as the playback rate.

Abstract: In one embodiment, a channel in a frame of the audio signal is subdivided into a plurality of blocks according to a subdivision hierarchy. The subdivision hierarchy has more than one level, and each level being associated with a different block length. At least two of the blocks have different lengths. An optimum prediction order is determined for each block based on a maximum permitted prediction order and a length of the block, where a prediction order is the number of linear prediction coefficients. The optimum prediction order is selected as a minimum one of the global prediction order and the local prediction order. The global prediction order is determined based on the maximum permitted prediction order, and the local prediction order is determined based on the length of the block.

Abstract: To increase channel capacity, mobile phone carriers have deployed speech coders, such as Advanced MultiBand Excitation coding (AMBE), in networks to reduce the bit rate of each call. One undesired consequence of employing such speech coders is that the voice quality can be much worse as compared to higher bit-rate speech coders. A method or corresponding apparatus in an example embodiment of the present invention performs voice quality enhancement transparently within a network by detecting use of a coder applying rate reduction to a speech signal and known to have an adverse effect on a coded speech signal. Upon detection of the use of such coder, the coded speech signal is corrected based on components introduced into the coded speech signal due to the rate reduction. As a result of applying the voice quality enhancement, adverse effects of speech coders can be reduced, while maintaining high quality voice signals.

Abstract: Encoding audio signals for Discontinuous with selecting an encoding mode for encoding the signal categorizing the signal into active segments having voice activity and non-active segments having substantially no voice activity by using categorization parameters depending on the quality of the selected encoding mode and encoding at least the active segments using the selected encoding mode that for a low quality encoding produce a lower number of “active” temporal section detections than for a high quality encoding mode, with comfort noise parameters producing less contrast from background noise for low quality encoding than for high quality modes.

Abstract: Systems and methods provide security intelligence knowledge of network communications. Particularly, VoIP packetized data being communicated over the network, together with other types of data communications over the network, is monitored and detected. From the VoIP packetized data, select portions or streams of the VoIP packetized data are identified. The select packetized data is then matched to a fingerprint or other indicia that indicates that the data is suspect for security intelligence knowledge purposes. The matched data is translated for purposes of taking security precautions based on the content of the security intelligence knowledge so gleaned.