Abstract: A speech recognition system recognizes an input speech signal by using a first speech recognizer and a second speech recognizer each coupled to a decision module. Each of the first and second speech recognizers outputs first and second recognized speech texts and first and second associated confidence scores, respectively, and the decision module selects either the first or the second speech text depending upon which of the first or second confidence score is higher. The decision module may also adjust the first and second confidence scores to generate first and second adjusted confidence scores, respectively, and select either the first or second speech text depending upon which of the first or second adjusted confidence scores is higher. The first and second confidence scores may be adjusted based upon the location of a speaker, the identity or accent of the speaker, the context of the speech, and the like.

Abstract: The telephone apparatus of the present invention comprises a first voice band expander for generating a voiced signal frequency component by shifting the frequency of the voice signal received, a second voice band expander for generating a voiceless signal frequency component by shifting the frequency of the voice signal received, and a voice composer for composing the voice signal received, the output of the first voice band expander, and the output of the second voice band expander, which is able to output clear voices in aural communication.

Abstract: A signature array of digitally sampled audio is formed from segment(s) of the digitally sampled audio by counting the number of occurrences within the segment(s) in each of a plurality of value bands or slots, such as amplitude bands. The signature array undergoes a fuzzy comparison with signatures arrays in the database. If more than one potential match is found, a more precise comparison is made. In the case of compact discs (CDs), five second sample segments may taken from the beginning, middle and end of each track to detect, e.g., the amplitude of the digitally sampled audio on the CD. A CD signature array may be formed of approximately 2000 value bands or slots by accumulating the occurrence of signals within each slot for all of the sample segments of the CD.

Abstract: The invention relates to a method and to a device for identifying markers in data blocks. The method is characterized in that a number of data blocks (201a 201n) are received by a receiver (202) and one specific data block from the number of data blocks (201a 201n) is analyzed in order to determine whether the specific data block contains a marker (203) or not. Once a data block that contains a marker (203) is identified, a full-rate signaling block (206) that is transmitted at full rate and a half-rate signaling block (207) that is transmitted at half rate are searched for markers. A predetermined reference pattern is provided for correlation that is subsequently divided into sub-ranges that in turn are correlated with a predetermined reference pattern to identify a signaling frame in the full-rate signaling blocks (206) and/or the half-rate signaling blocks (207).

Abstract: There are provided a system and method for providing information using a spoken dialogue interface. The system includes a speech recognizer for transforming voice signals into sentences; a sentence analyzer for analyzing the sentences by their structural elements; a dialogue manager for extracting information on speech acts or intentions from the structural elements, and generating information on system's speech acts or intentions for a response to the extracted information on speech acts or intentions; a sentence generator for generating sentences based on the information on the system's speech acts or intentions for the response; a speech synthesizer for synthesizing the generated sentences into voices; an information extractor for extracting information required for the response from the Internet in real time; and a user modeling means for analyzing and classifying users' tendencies.

Abstract: An automatic interpreting system, having at least an inputting unit for inputting utterances, an interpreting unit for recognizing the input utterance and interpreting the input utterance into a specified language, and an outputting unit for outputting interpretation results, includes a conversation history managing unit for determining the level of interpretation reliability calculated based on the interpretation result, having the user input their level of interpretation understanding, and recording and storing this data as interpretation history information, and a conversation status determination processing unit for, if the interpretation of an utterance is not understandable to the receiving party, determining the conversation status by selecting from categories pre-determined based on the levels of interpretation reliability and interpretation understanding for the previous utterance that are stored in the conversation history managing unit, and a recommended action presenting part for presenting to the

Abstract: This invention is directed to a method of delivering vehicle owner's manual or other vehicle-specific information to the vehicle operator from a remote data center and associated vehicle information database by utilizing a voice recognition system at the remote data center and delivering the information to the vehicle operator in audible speech. The vehicle operator speaks his request in the vehicle and the data center recognizes the request, perhaps asks more questions, leads the vehicle operator through a spoken menu, and then provides the answer vocally to the vehicle operator over the speaker(s) located in the vehicle. The invention includes methodology for obtaining vehicle diagnostic information and controlling certain vehicle functions automatically via an embedded telematics control unit. The invention further includes remote telephone access outside the vehicle.

Abstract: To provide a robot which autonomously forms and performs an action plan in response to external factors without direct command input from an operator. When reading a story printed in a book or other print media or recorded in recording media or when reading a story downloaded through a network, the robot does not simply read every single word as it is written. Instead, the robot uses external factors, such as a change of time, a change of season, or a change in a user's mood, and dynamically alters the story as long as the changed contents are substantially the same as the original contents. As a result, the robot can read aloud the story whose contents would differ every time the story is read.

Abstract: A method for transmitting a character message using voice recognition in a portable terminal, includes the steps of: a) displaying a guidance message associated with a voice/character conversion service when the message is created; b) connecting the portable terminal to a base station when the voice/character conversion service is selected; c) transmitting a voice message from a user to the base station; d) storing the voice message in the base station; e) allowing the base station to convert the voice message into a character message; f) transmitting the character message to the portable terminal; g) inputting a destination number; and h) transmitting the character message to a destination terminal corresponding to the inputted destination number.

Abstract: A wireless web system allows users to navigate web pages that include links to audio content where the pages are provided over a data connection and the audio content is provided over a voice connection. An audio content reference generator generates a reference to a portion of static audio content and that audio content reference is provided to the user's wireless web client as a link on a wireless web page, or other page retrieved by the wireless web device over the data connection. The audio content reference and a telephone number of an audio server form the link on the page, so that when a user selects that link, the wireless device establishes a voice connection to the audio server using the telephone number and then provides the audio server with the audio content reference so that the user hears the specifically referenced audio content over the voice channel.

Abstract: The quality of audiovisual material is assessed by measuring the audio an video quality and computing from these a combined measure. Using a parameter indicative of the degree of motion represented by the video, the computation employs one of a plurality of algorithms selected in dependence on the value of the parameter.

Abstract: A word-device setting information storage part stores multiple device setting information in association with a single voice and a voice recognition part recognizes a voice input through a voice input part so that the multiple device setting information associated with the recognized voice is read from the word-device setting information storage part to a device control part, which in turn sets an internal device state of a device setting apparatus while an external device control part sets a state of an external device in response to the read multiple device setting information.

Abstract: An electronic device which includes a user interface having a display for displaying text and a speech synthesiser including a loudspeaker, arranged to convert an input, dependent upon a text, to an audio output representative of a person reading the text. The device may delay with respect to the audio output corresponding to a portion of text the display of the portion of text. The device may also or alternatively delay with respect to the audio output corresponding to a portion of text the highlighting of the portion of text within a displayed text.

Abstract: An adaptive noise suppression system includes an input A/D converter, an analyzer, a filter, and a output D/A converter. The analyzer includes both feed-forward and feedback signal paths that allow it to compute a filtering coefficient, which is input to the filter. In these paths, feed-forward signal are processed by a signal to noise ratio estimator, a normalized coherence estimator, and a coherence mask. Also, feedback signals are processed by a auditory mask estimator. These two signal paths are coupled together via a noise suppression filter estimator. A method according to the present invention includes active signal processing to preserve speech-like signals and suppress incoherent noise signals. After a signal is processed in the feed-forward and feedback paths, the noise suppression filter estimator then outputs a filtering coefficient signal to the filter for filtering the noise out of the speech and noise digital signal.

Abstract: Square sum calculator 603 calculates a square sum of evolution in smoothed quantized LSP parameter for each order. A first dynamic parameter is thereby obtained. Square sum calculator 605 calculates a square sum using a square value of each order. The square sum is a second dynamic parameter. Maximum value calculator 606 selects a maximum value from among square values for each order. The maximum value is a third dynamic parameter. The first to third dynamic parameters are output to mode determiner 607, which determines a speech mode by judging the parameters with respective thresholds to output mode information.

Abstract: A method for reducing noise in a voice signal, and a voice operated system utilizing the same are presented. A noise component in a compressed digital signal representative of the voice signal is determined, and subtracted from the compressed digital signal.

Abstract: A method for multi-modal data fusion (100), a multi-modal system data fusion system (10) and module (24) that in use operates by receiving segments (125) of multi-modal data associated respectively with a modalitiy. Initiating (130) a dynamically variable wait period after one of the segments is received is then performed. The dynamically variable wait period has a duration determined from data fusion timing statistics of the system (10). A waiting (140) for reception of any further segments during the dynamically variable wait period is then effected and thereafter a fusing (145) of the segments received provides fused data that is sent (160) to a dialog manager (25).

Abstract: A speech recognition method includes a step of receiving a phonetic sequence output by a phonetic recognizer. The method also includes a step of matching the phonetic sequence with one of a plurality of reference phoneme sequences stored in a reference list that matches closest thereto. At least one of the plurality of reference phoneme sequences stored in the reference list includes additional information with respect to a phonetic sequence that is capable of being output by the phonetic recognizer.

Abstract: A data-compressed audio waveform is temporally modified without requiring complete decompression of the audio signal. Packets of compressed audio data are first unpacked, to remove scaling that was applied in the formation of the packets. The unpacked data is then temporally modified, using one of a number of different approaches. This modification takes place while the audio information remains in a data-compressed format. New packets are then assembled from the modified data, to produce a data-compressed output stream that can be subsequently processed in a conventional manner to reproduce the desired sound. The assembly of the new packets employs a technique for inferring an auditory model from the original packets, to requantize the data in the output packets.

Abstract: Words are spelled by receiving recognizable words from a user of an interactive voice response system. The first letter of each recognizable word is identified, and a spelling is determined based on the first letters of the recognizable words. Statistics for previous users of the interactive voice response system are determined, where the statistics indicate the number of times each of the recognizable words has been used to indicate a letter. The recognizable word that is most commonly used for each letter is identified. The user is prompted with at least two recognizable words that are most commonly used, where each recognizable word corresponds to a different letter. A selection of one of the recognizable words provided to the user is received.