A band scrambler which processes only time domain samples is described. The band scrambler has the effect of dividing the input signal spectrum into N sub-bands. The N sub-bands are permuted such that the r th band is mapped onto the k.r th band modulo N, where N is a constant of the scrambler and k is the key which is variable in the range 2<k<N-1. The output samples y(n) produced by the scrambler from the input speech signal samples x(n) are defined by the equation: ##EQU1## The down-sam...

Methods and systems, including computer program products, for speech animation. The system includes a speech animation engine and a client application in communication with the speech animation engine. The client application sends a request for speech animation to the speech animation engine. The request identifies data to be used to generate the speech animation, where speech animation is speech synchronized with facial expressions. The client application receives a response from the speech ani...

A method of speech analysis calculates one or more difference parameters for each of a sequence of acoustic frames, where each difference parameter is a function of the difference between an acoustic parameter in one frame and an acoustic parameter in a nearby frame. The method is used in speech recognition which compares the difference parameters of each frame against acoustic models representing speech units, where each speech-unit model has a model of the difference parameters associated with...

Techniques for assisting in translation are provided. A speech recognition hypothesis is obtained, corresponding to a source language utterance. Information retrieval is performed on a supplemental database, based on a situational context, to obtain at least one word string that is related to the source language utterance. The speech recognition hypothesis and the word string are then formatted for display to a user, to facilitate an appropriate selection by the user for translation.

A preferred embodiment of the method for converting text to speech using a computing device having a memory is disclosed. The inventive method comprises examining a text to be spoken to an audience for a specific communications purpose, followed by marking-up the text according to a phonetic markup systems such as the Lessac System pronunciation rules notations. A set of rules to control a speech to text generator based on speech principles, such as Lessac principles. Such rules are of the tide ...

The generic word label series used for recognition of words uttered by unspecified speakers are stored in the vocabulary label network accumulation processing. The speech of a particular speaker is entered. Based on the input speech, the registered word label series extraction processing generates the registered word label series. The registered word label series of the particular speaker can then be registered with the vocabulary label network accumulation processing.

Portions of recorded speech waveform (e.g., corresponding to phonemes) are combined to synthesize words. In order to provide a smoother delivery, each voiced portion of a waveform portion has its amplitude adjusted to a predetermined reference level. The scaling factor used is varied gradually over a transition region between such portions and between voiced and unvoiced portions.

A decoding processing portion 11 of a speech decoder 10 is provided with an emphasis processing portion 15 for performing an emphasis process on signals to be processed (excited signals) SPC generated from coded speech signals BS. A counter portion 17 counts the number of times code errors occurred in successive frames of the coded speech signal BS, and outputs the successive frame error number. When the successive frame error number outputted form the counter portion 17 is less than or equal to...

A variable bit-rate speech coding method determines for each subframe a quantised vector d(i) comprising a variable number of pulses. An excitation vector c(i) for exciting LTP and LPC synthesis filters is derived by filtering the quantised vector d(i), and a gain value g.sub.c is determined for scaling the pulse amplitude excitation vector c(i) such that the scaled excitation vector represents the weighted residual signal s remaining in the subframe speech signal after removal of redundant info...

A method of speech training including sensing speech of a student and providing a digital indication of the sensed speech. Apparatus in support of this method is additionally provided.