Abstract: A computer telephone relay service for interfacing between a textphone and a telephone. A textphone call is made to the computer which makes a further call to a telephone or vice versa. The computer then translates the messages between the text-phone and telephone using voice recognition and text-to-speech hardware.

Abstract: A transcoder which prevents tandem coding of speech in a mobile-to-mobile call within a mobile communication system uses a speech coding method for reducing transmission rate on the radio path. The transcoder includes a speech coder, which encodes the speech signal into speech parameters for transmission to a mobile station, and decodes the speech parameters received from the mobile station into a speech signal according to the speech coding method, as well as a PCM coder for transmitting an uplink speech signal to and for receiving a downlink speech signal from a PCM interface in the form of PCM speech samples. In addition to the normal operation, the transcoder transmits and receives speech parameters through a PCM interface in a subchannel formed by least significant bits of the PCM speech samples. Thus, it is possible to prevent tandem coding while maintaining the standard PCM interface, and the signaling and services associated thereto.

Abstract: Methods and systems consistent with the present invention enroll a candidate phrase uttered by a user in a dictionary having at least one previously enrolled phrase. The system receives utterances of the candidate phrase and determines whether the first utterance is confusingly similar to a previously enrolled phrase and whether they are consistent with each other. The system then enrolls the candidate phrase in the dictionary according to these determinations.

Abstract: A DVD audio disk having information areas in which corresponding audio title information management tables are stored, and corresponding data areas in which audio packs of a linear PCM mode are stored, wherein a first, second or third number of quantization bits, a corresponding first, second or third sampling frequency, and information relative to the number of audio channels are all recorded on each audio title information management table. The audio pack includes audio packets made up with the number of quantization bits, the sampling frequency and the information relative to the number of channels all recorded on the title management table, the audio packets further containing audio data.

Abstract: The object of the invention is to provide a coding/decoding method in which degradation of sound quality perceptible by the listener does not occur at an low bit rate. A shift number calculation section of a decoding device divides a frequency domain into at least two sub-bands, and approximates each of normalized transform coefficients in the sub-band whose allocated bit value is less than a predetermined threshold using a quantized value of the transform coefficient in a predetermined sub-band other than the sub-band so as to obtain information concerning the approximation, and a multiplexer multiplexes the information and another signal and transmits them. A de-multiplexer of a decoding device separates the code of information concerning the approximation, and a shift number restore section restores the information based thereon.

Abstract: A method and apparatus for automatically compensating a tone color. The method includes a first step for determining whether a channel is modified or an input audio signal is switched, a second step for calculating frequency characteristics of the input audio signal and comparing the compared result with data in a basic table, a third step for determining the input audio signal as a least error mode during the second step, and a fourth step for compensating a tone color in accordance with the determined mode. The method and apparatus automatically adjusts a tone color by determining a present tone color in itself on the ground of the frequency energy being presently applied when a channel is modified, an input audio signal is switched, or a color sound mode is modified, thereby eliminating a manual manipulation by a user as well as providing an optimal sound.

Abstract: A speaker clustering apparatus generates HMMs for clusters based on feature quantities of a vocal-tract configuration of speech waveform data, and a speech recognition apparatus provided with the speaker clustering apparatus. In response to the speech waveform data of N speakers, an estimator estimates feature quantities of vocal-tract configurations, with reference to correspondence between vocal-tract configuration parameters and Formant frequencies predetermined based on a predetermined vocal tract model of a standard speaker. Further, a clustering processor calculates speaker-to-speaker distances between the N speakers based on the feature quantities of the vocal-tract configurations of the N speakers as estimated, and clusters the vocal-tract configurations of the N speakers using a clustering algorithm based on calculated speaker-to-speaker distances, thereby generating K clusters.

Abstract: A method for reducing noise in an input speech signal by adaptively controlling a maximum likelihood filter that is provided to calculate speech components based on a probability of speech occurrence and on a calculated signal-to-noise ratio based on the input speech signal. The characteristics of the maximum likelihood filter are smoothed along both the frequency axis and along the time axis. In the case of the frequency axis, smoothing filtering is based upon a median value of characteristics of the filter in the frequency range under consideration and on the characteristics of the filter in neighboring left and right frequency ranges, and in the case of smoothing filtering along the time axis, smoothing is done both for signals of a speech part and of a noise part.

Abstract: An ACELP speech coding method according to ITU-T Recommendation G.729. When coding a random component vector, each of random component vector forming together the random codebook is formed of three or less pulses having a unit amplitude for each 6f a pair of subframes which form together a frame. The positions of the pulses are determined from a plurality of predetermined positions which a pulse can assume in a subframe so that distortion is minimized. The method allows speech coding at a lower bit rate.

Abstract: A voice converter provides for pitch and formant shifting of an input voice signal. An audio filter extracts the volume level of the input voice signal, and outputs the extracted volume level as first volume data. A second audio filter extracts the volume level of an output voice signal, and outputs the extracted volume level as second volume data. A difference judging circuit compares the first and second volume data with each other, and determines a volume gain and a distorting factor which is supplied to a distortion circuit. When the volume of the output voice after conversion is smaller than that of the input voice, the volume gain is increased.

Abstract: A speech processing apparatus includes a noise model production device for extracting a noise-speech interval from input speech data and producing a noise model by using the data of the extracted interval. The apparatus also includes a composite distribution production device for dividing the distributions of a speech model into a plurality of groups, producing a composite distribution of each group, and determining the positional relationship of each distribution within each group. In addition, the apparatus includes a memory for storing each composite distribution and the positional relationship of each distribution within the group, and a PMC conversion device for PMC-converting each produced composite distribution. Also provided is a noise-adaptive speech model production device for producing a noise-adaptive speech model on the basis of the composite distribution which is PMC-converted by the PMC conversion device and the positional relationship stored by the memory.

Abstract: A digital speech communication system having improved synchronization. The present digital speech communication system reduces the unit of degradation to a single speech sample, rather than a multi-sample frame, while maintaining the bit rate efficiency of the DSVD system and other systems where speech is encoded into large blocks and is subject to variable delay and mismatched clocks. The basic unit that is dropped or artificially inserted by the receiver, if the buffer overflows or empties, respectively, is reduced to a single speech sample. The speech frames produced by the demultiplexer are written into a frame buffer, in units of frames, at a rate determined by the clock signal, S2, that is extracted from the received signal by a timing recovery function in the modem. In accordance with the present invention, the frames are read out of the buffer into the decoder using the same extracted clock signal, S2. In this manner, once the buffer is partially full, the frame buffer will not overflow or empty.

Abstract: A composite pitch pattern of an artificial waveform of a composite sound indicating characters is produced according to a general pitch pattern producing model, and a pitch pattern of a VCV phoneme-chain waveform of each of VCV phoneme-chains corresponding to the characters is produced from an actual voice sample. Each VCV phoneme-chain composed of a preceding vowel, a consonant and a succeeding vowel has a pitch fine structure and a pitch fluctuation. Thereafter, an overall inclination of the pitch pattern of each VCV phoneme-chain waveform is adjusted to that of a portion of the composite pitch pattern corresponding to the-same VCV phoneme-chain to overlap transitional portions of preceding and succeeding vowels in a changed pitch pattern of each VCV phoneme-chain waveform with those in the corresponding portion of the composite pitch pattern.

Abstract: In a speech reproducing system, a speech coder receives an input speech signal to output a speech coded information including a pitch information of the input speech signal and a mode information indicative of a short-time characteristics of the input speech signal, and a speech decoder receives and decodes the speech coded information to generate a decoded speech signal. A speech-rate converter receives the pitch information and the mode information included in the speech coded information and the decoded speech signal, to convert the speech-rate of the decoded speech signal by using the pitch information and the mode information, thereby to generate an output speech signal.

Abstract: An audio interactive system including a plurality of terminals each having an audio input means for converting voices into aural signals and an audio output means for converting aural signals into voices, the plurality of terminals being connected to a communication line; and a server connected with the plurality of terminals via the communication line to perform collection and distribution of the aural signals, in which a conversation is made through voices between the plurality of terminals. The server comprises a buffer for temporarily storing the aural signals transmitted from the terminals, and a scheduler for controlling the distribution of the aural signals stored in the buffer.

Abstract: A method for automatically confirming verbal commands issued by a user and received by an automatic speech recognition system. Each verbal command is associated with a command operator set in which one or more command operators are stored. After the user vocalizes a recognized verbal command, the automatic speech recognition system awaits vocalization of a valid command operator from an associated stored command operator set for a pre-determined period of time. During this waiting period a non-intrusive indicator such as a light or a low volume tone notifies the user that the automatic speech recognition system is awaiting or ready to receive an appropriate command operator. If the user vocalizes a valid and recognized command operator during the waiting period, then the automatic speech recognition system executes the verbal command using the command operator and issues a confirmation to the user that the verbal command has been executed.

Abstract: A control system for aiding the operator of a vehicle or platform, such as an aircraft. The system includes a receiver adapted to receive a voice command from a controller located remotely from the vehicle or platform. A speech recognition device is coupled to the receiver and adapted to compare the voice command to speech recognition templates of known voice commands. A match between the voice command and a speech recognition template is indicative of informational content of the voice command. A display device coupled to the speech recognition device displays to the operator of the vehicle or platform a visual representation of the informational content of the voice command.

Abstract: A method for selecting a telephone number by means of voice control, the telephone numbers which can be selected are stored, and for each telephone number which can be selected at least one identifier, such as a name is stored. In the storing phase, the identifier is pronounced and divided into one or several sub-identifiers, which are stored, and to which information on the telephone number is linked, wherein the telephone number mentioned in the selecting phase can be dialed either by pronouncing said sub-identifiers in any order according to a combination or partial combination of the sub-identifiers.

Abstract: A method and apparatus for the selection of an encoding mode for speech frames in a variable rate encoding system. For each speech frame, the method and apparatus selects the encoding mode which provides for rate efficient coding. A mode measurement element receives a speech signal and a signal derived from the same speech signal, and generates a set of parameters which are ideally suited for operational mode selection. Rate determination logic receives the set of parameters and selects an encoding rate using predetermined selection rules. The selection rules further distinguish between unvoiced speech and temporally masked speech, which are encoded at the same rate but with different encoding strategies.

Abstract: A computerized speech recognition system creates acoustic models of phrases by concatenating acoustic models for individual words. The system stores an acoustic word model and spelling for each of its vocabulary words. When it receives the spelling of a multi-word phrase to be treated as a new vocabulary word, it stores that multi-word spelling as the spelling of the new vocabulary word, and a new acoustic model created by concatenating the acoustic word models of previous vocabulary words whose spellings correspond to words in the multi-word spelling as the acoustic model for the new word. The system can then perform speech recognition by comparing acoustic signals against the word models of stored vocabulary words, including those representing such multi-word phrases. Preferably when a multi-word model is formed, the individual acoustic models concatenated are modified to represent the coarticulation which takes place between words spoken continuously.