Abstract: In a method of and an apparatus for the segmentation of speech, an acoustic speech signal is converted into N signals S.sub.i (f), each signal pertaining to a time interval i of N successive time intervals (1.ltoreq.i.ltoreq.N). For successive time intervals i a function c.sub.ij (FIG. 7) is then derived which is a measure of the agreement between the signals in the time intervals i and j. The middle m.sub.i of the pertaining function C.sub.ij is determined for each time interval i and those values i are determined which at least approximately correspond to zero-axis crossings with the same sign in a function d.sub.i (see FIG. 3b), d.sub.i indicating the difference between m.sub.i and i. The signal segmented in this way can be used for the derivation of diphones so that a library of diphones can be built up (see FIG. 1), or it can be used for the recognition of the speech utterance (see FIG. 2).

Abstract: There is provided an automatic voice recognition system which utilizes time encoded speech. Through the determination of zero crossing information and waveform parameters of an input voice signal, a stream of time encoded speech symbols is obtained. The stream of time encoded speech symbols is then converted into a matrix format for comparison with reference matrices formatted from time encoded symbols of reference words thereby to provide an output signal indicative of the content of the input voice signal.

Abstract: A pattern recognizer uses the method of n-tuples applied to a matrix containing (e.g.) features derived from input speech. A template store records during a training sequence the frequencies of occurrence of combinations of bits with each n-tuple group. During recognition frequencies associated with combinations which occur are used to form a measure of similarity.

Abstract: A phoneme estimator (12) in a speech-recognition system (10) includes trigger circuitry (18, 22) for identifying the segments of speech that should be analyzed for phoneme content. Speech-element processors (24, 26, and 28) calculate the likelihoods that currently received speech contains individual phonemes, but they operate only when the trigger circuitry identifies such segments. The computation-intensive processing for determining phoneme likelihoods is thus performed on only a small subset of the received speech segments. The accuracy of the speech-element processors (24, 26, and 28) is enhanced because these processors operate by recognition of patterns not only in elements of the data-reduced representations of the received speech but also in higher-ordered products of those elements; that is, these circuits employ non-linear modeling for phoneme identification.

Abstract: A communication system having an encoder device used in combination with a decoder device for encoding a sequence of digital speech signals into a sequence of output signals, using a spectrum parameter and a pitch parameter. A subsidiary parameter of the digital speech signals is detected and monitored by a monitoring circuit. Digital speech signals are classified into voiced sound or voiceless sound and into vocality, nasal, fricative, or explosive durations at every frame. When a voiced sound, i.e., a vocality is detected, a predetermined number of excitation pulses are calculated during a representative subframe and are produced as primary sound source signals. A subsidiary information signal is produced during the remaining subframes to represent phase and amplitude correction factors in each of the subframes. When a voiceless sound, i.e.

Abstract: Apparatus for order entry in a restaurant includes a host and remote portable terminals. Orders are entered on the remote terminals and sent by radio frequency communication to the host systems. The portable terminals include a bar code template of menu items and a light pen for scanning menu item selections. Orders are entered by scanning selections in a manner similar to the use of a conventional restaurant order pad. Terminals and printers are connected to the host system and are used to display orders entered at portable terminals for the cooks in the kitchen and to print order receipts.

Abstract: When a speech sound of at least a predetermined sound pressure is externally input while a time measurement is not being performed, a time measuring circuit starts a time measurement responsive to a signal from a speech detector. When another speech sound of at least a predetermined sound pressure is externally input while a time measurement is being performed by the time measuring circuit, a measurement time measured by the time measuring circuit at this moment is stored in a time information memory. After a predetermined time has elapsed, if a speech recognition circuit recognizes that the externally input speech sound is a "stop" command, the time measurement operation performed by the time measuring circuit is stopped, and the time information stored in the time information memory is read out and displayed as measurement time information on a display unit.

Abstract: A circuit arrangement for storing a speech signal in a digital speech memory (SP) and reproducing the stored signal comprises among other things a control circuit (SS). The control circuit produces control commands and clock signals for the individual modules of the arrangement. The speech signal to be stored is available in one of various predetermined codes. In order to enable storing speech signals of considerable lengths, it is provided to write the speech signal into the digital speech memory (SS) in a preferred digital code. For this purpose, input converters (EW1, EW2, EW3, EW4) are used which convert the code of the speech signal to be stored from the one it has into the preferred digital code, if the signal is not already available in this code. Multiplexing means (M1, M2), controlled by the control circuit (SS), further convey the converted speech signal to the input of the digital speech memory (SP).

Abstract: An echo effect is synthesized in a coded digital sound signal. An input sound signal is stored as a plurality of sequential frames of similar duration, each frame (n) having characteristics including an energy (E). A delay period (d) is selected as equal to a number of time. For each frame (n) later than the duration of frames of the delay (d), the energy E(n) of the frame is compared to an attenuated energy aE(n-d) of an earlier frame (n-d), which is earlier in time than frame (n) by a number of frames equal to the delay (d). If the energy E(n) is less than the attenuated energy aE(n-d) of the earlier frame, the current frame is replaced in an output sequence with a new frame having the non-energy characteristics of the earlier frame and the attenuated energy.

Abstract: Apparatus for use by a broker including a memory for holding a number of market prices each corresponding to a security or a commodity. A speech synthesizer provides that prices may be quoted to customers in a synthesized human voice. A telephone interface connects outside calls to a DTMF transceiver which is in turn connected to a programmed microprocessor unit so that a caller may key in instructions or information. Instructions for price quotations result in the price for the selected security or commodity being reported to the caller in a synthesized human voice. The apparatus is also capable of monitoring price triggers and reporting violations of price triggers by calling a predetermined telephone number and reporting the violation in a synthesized human voice. Also, price requests are logged and the unit is responsive to an instruction keyed in by a broker through his telephone to report the contents of the log over the telephone in a synthesized human voice.

Abstract: The input pattern to a pattern recognition system is defined as a candidate graph composed of several possibly valid category strings. The most valid category string is selected as the input pattern. Each category string is composed of linked category segments, each category segment being bounded by category nodes or branching points to which are connected one or more category segments. The most valid category string is developed by establishing the most valid midway path from the start node and extending to the terminal node. A most valid midway path to each midway node (n) is established by using the minimum valid estimation value determined for all possible midway paths from the start node to the midway node (n).

Abstract: In circuits for generating rapidly changing, yet smooth and controlled, control signals in response to transient signals for use in audio compressors and expanders, a steady-state portion of the control circuit includes first and second smoothing stages (alternatively, a single stage may be employed) and an overshoot suppression portion of the control circuit rapidly charges the capacitor in the steady-state smoothing stage from which the control voltage is derived. A first additional overshoot suppression circuit provides further charging of the capacitor in response to low frequency signal components. Modulation control signals, which make the compressor or expander less susceptible to control by stop-band signal components, are combined in opposition to the signals in the steady-state and overshoot suppression portions of the control circuit.

Abstract: A speech synthesis system comprises an input terminal for accepting text code, accent code, and phrase code.

Abstract: A pitch detector system for use with speech analysis and synthesis methods having a plurality of identical detectors each responsive to a different portion of a speech signal for estimating a pitch value and a voter circuit responsive to the estimated pitch values for determining a final pitch value. The pitch detectors are identical in design which allows for an efficient software implementation since only one set of program instructions is necessary to implement all of the encoders. The voter subsystem may be implemented by a digital signal processor executing program instructions that calculate a pitch value from the estimated pitch values determined by the pitch detectors and a second set of program instructions for constraining the final pitch value outputted by the voter subsystem so that the calculated pitch value is in agreement with calculated pitch values for previous frames.