Abstract: The invention is embodied in a computer user interface including an observer capable of observing user behavior, an agent capable of conveying emotion and personality by exhibiting corresponding behavior to a user, and a network linking user behavior observed by said observer and emotion and personality conveyed by said agent. The network can include an observing network facilitating inferencing user emotional and personality states from the behavior observed by the observer as well as an agent network facilitating inferencing of agent behavior from emotion and personality states to be conveyed by the agent. In addition, a policy module can dictate to the agent network desired emotion and personality states to be conveyed by the agent based upon user emotion and personality states inferred by the observing network. Typically, each network is a stochastic model.

Abstract: A multiple confidence measures subsystem of an automated speech recognition system allows otherwise independent confidence measures to be integrated and used for both training and testing on a consistent basis. Speech to be recognized is input to a speech recognizer and a recognition verifier of the multiple confidence measures subsystem. The speech recognizer generates one or more confidence measures. The speech recognizer preferably generates a misclassification error (MCE) distance as one of the confidence measures. The recognized speech output by the speech recognizer is input to the recognition verifier, which outputs one or more confidence measures. The recognition verifier preferably outputs a misverification error (MVE) distance as one of the confidence measures. The confidence measures output by the speech recognizer and the recognition verifier are normalized and then input to an integrator.

Abstract: For use in a synthesizer having a wave source that produces a periodic wave, frequency shifting circuitry for frequency-shifting the periodic wave and waveshaping circuitry for transforming the periodic wave into a waveform containing a formant, the frequency-shifting causing displacement of the formant, a circuit for, and method of, compensating for the displacement and a synthesizer employing the circuit or the method. In one embodiment, the circuit includes bias circuitry, coupled to the wave source and the frequency shifting circuitry, that introduces a bias into the periodic wave based on a degree to which the frequency shifting circuitry frequency shifts the periodic wave, the bias reducing a degree to which the formant is correspondingly frequency-shifted.

Abstract: A voice response apparatus and method in which narrative text contained in a database is presented to a user through a telephone. Based on user responses, the voice response apparatus selects only the appropriate text which matches the user's selection. The user has the option of listening to a human voice synthesized by the system reciting the text or having the text and corresponding graphics faxed to him. At any point during the recitation of text, the user may select certain options made available by the system. These options include among many: increasing the speed of the voice reciting the text; decreasing the speed of the voice reciting the text; listening to a summary of the text rather than the full text; discontinuing recitation of the text; and switching to a different text. The system, upon detection of a user option selection, marks a position in the text and continues the recitation from that point when appropriate depending on the option selected.

Abstract: A method and system for providing user access to complex information through interactive voice dialog. The computerized method includes the provision of a memory and storage in the memory of a plurality of selected words which may be recognized in predetermined phrases and dialogue contexts when spoken by users. A voice template is further stored in memory having selected information slots in corresponding frames. Each of the slots and frames is adapted to be continuously filled by recognized words. User speech utterances are initiated for receipt by the computer which requests complex information or responds to generated queries. By continuously receiving and comparing the user speech utterances to the stored plurality of selected words, slots and frames of the voice template are filled.

Abstract: Disclosed is a method for drastically reducing the average error rate for signals under mismatched conditions. The method takes a signal (e.g., speech signal) and a set of stored representations (e.g., stored representations of keywords) and performs at least one transformation that results in the signal more closely emulating the stored representations. This is accomplished by using one of three techniques. First, one may transform the signal so that the signal may be better approximated by (e.g., is closer to) one of the stored representations. Second, one may transform the set of stored representations so that one of the stored representations better approximates the signal. Third, one may transform both the signal and the set of stored representations.

Abstract: An improved long-term predictor (LTP) for use in analysis-by-synthesis coding systems, such as CELP is disclosed. The invention provides control of the periodicity of speech signals generated by the LTP. This control facilitates a reduction in perceptible noise/buzziness in reconstructed speech. An embodiment of the invention includes a conventional LTP in combination with a two-tap finite impulse response filter. The filter augments operation of the LTP by generating precursor signals of LTP output signals. These precursor signals are combined with the LTP output signals to form the output of the improved LTP.

Abstract: A text-to-speech system includes a memory storing a set of quantization vectors. A first processing module is responsive to the sound segment codes generated in response to text in the sequence to identify strings of noise compensated quantization vectors for respective sound segment codes in the sequence. A decoder generates a speech data sequence in response to the strings of quantization vectors. An audio transducer is coupled to the processing modules, and generates sound in response to the speech data sequence. The quantization vectors represent a quantization of a sound segment data having a pre-emphasis to de-correlate the sound samples used for quantization and the quantization noise. In decompressing the sound segment data, an inverse linear prediction filter is applied to the identified strings of quantization vectors to reverse the pre-emphasis. Also, the quantization vectors represent quantization of results of pitch filtering of sound segment data.

Abstract: A telecommunications network service overcomes the annoying effects of transmitted noise by a signal processing which filters out the noise using interactive estimations of a linear predictive coating speech model. The speech model filter uses an accurate updated estimate of the current noise power spectral density, based upon incoming signal frame samples which are determined by a voice activity detector to be noise-only frames. A novel method of calculating the incoming signal using the linear predictive coating model provides for making intraframe iterations of the present frame based upon a selected number of recent past frames and up to two future frames. The processing is effective notwithstanding that the noise signal is not ascertainable from its source.

Abstract: A noise reducing method and device for reducing the noise contained in an input speech signal collects the speech signal with a microphone 11 and converts the speech signal into a digital input signal x(n) with an A/D converter 12. A frame power calculating circuit 13 calculates a mean frame power rms for each frame of the digital input signal x(n). A suppression ratio calculating circuit 14 calculates different values of the noise suppression ratio depending on the magnitude of the mean frame power rms relative to pre-set threshold values. A level discrimination circuit 18 forms a changeover control signal depending on the noise level and transmits the changeover control signal to the suppression ratio calculating circuit 14 for switching control of the threshold value.

Abstract: An apparatus for concealing a frame or multiple of frames where errors have occurred in a digital audio signal which is subband coded and transform coded in units of an error-correctable frame is described. The error concealment apparatus includes an error detector for receiving frequency coefficients representing the encoded digital audio signal and detecting whether an error has occurred for each frame, and a decoder for decoding the frequency coefficients by respective subbands to form a frequency domain of the whole audio signal with respect to the input frequency coefficients. A buffer stores the frequency coefficients decoded by said decoder. Frequency coefficients of a frame or multiple of frames where errors have occurred are reconstructed using predetermined weight values and frequency coefficients of adjacent frames which do not have errors.

Abstract: A synthesizer includes a controller which generates an address signal in response to a trigger code corresponding to a sequence of a synthesis of a plurality of basic speech sections; a memory for storing sets of data corresponding to the sequence of the synthesis of the speech sections; a tone counter and a speech/melody generator which receives the data from the memory. In response to control signals from the controller and a tone control signal from the tone counter the speech/melody generator provides synthesized speech or melody mixing with each other in a selective manner.

Abstract: A method and apparatus provide a video image of facial features synchronized with synthetic speech. Text input is transformed into a string of phonemes and timing data, which are transmitted to an image generation unit. At the same time, a string of synthetic speech samples is transmitted to an audio server. The audio server produces signals for an audio speaker, causing the audio signals to be continuously audibilized; additionally, the audio server initializes a timer. The image generation unit reads the timing data from the timer and, by consulting the phoneme and timing data, determines the position of the phoneme currently being audibilized. The image generation unit then calculates the facial configuration corresponding to the position in the string of phonemes, calculates the facial configuration, and causes the facial configuration to be displayed on a video device.

Abstract: The dictionary is broken into clusters by first grouping the dictionary according to a rule based procedure whereby the dictionary is sorted by word length and alphabetically. After sorting, a plurality of first cluster centers is generated by selecting the dictionary entries that differ from neighboring entries by the first letter. Each of the dictionary entries is then assigned to the closest one of the first cluster centers using a dynamic time warping procedure. These newly formed clusters are then each analyzed to find the true cluster center and the dictionary entries are then each assigned to the closest true cluster center. The clusters, so formed, may then be rapidly searched to locate any dictionary entry. The search is quite efficient because only the closest cluster to the desired dictionary entry needs to be searched.

Abstract: There is provided a speech efficient coding method applicable to, e.g., analysis by a synthesis system such as an MBE vocoder, and comprising the steps of (a) dividing an input speech signal into block units on a time base, (b) dividing signals of each of the respective divided blocks into signals in a plurality of frequency bands, (c) discriminating whether signals of each of the respective divided frequency bands which are lower than a first frequency are voiced sound or unvoiced sound, (d) if the discrimination results in step (c) for a predetermined number of frequency bands is voiced sound, assigning a discrimination result of voiced sound to all frequency bands lower than a second frequency which is higher than the first frequency to obtain an ultimate discrimination result of voiced sound/unvoiced sound.

Abstract: Azeotrope-like compositions comprising trichlorotrifluoroethane, ethanol, nitromethane and 2-methylpentane or a mixture of hexanes which are stable and have utility as degreasing agents and as solvents in a variety of industrial cleaning applications.