Motorola Corporation

Gerald Edward Corrigan

Orhan Karaali

Noel S Massey

Larry C Puhl

Ira A Gerson

Stephen N Levine

Harry M Bliss

The present invention provides a method, device and system to generate acoustic parameters in a text-to-speech system utilizing a neural network to generate a representation of a trajectory in an acoustic parameter space across a phonetic segment.

Method, apparatus and system for generating acoustic parameters in a text-to-speech system using a neural network

The present invention teaches a method (400), device and system (300) utilizing at least one of: mapping a sequence of phones to a sequence of articulatory features and utilizing prominence and boundary information, in addition to a predetermined set of rules for type, phonetic context, syntactic and prosodic context for phones to provide provide a system that generates segment durations efficiently with a small training set.

Method, device and system for generating segment durations in a

A method (400), device and system (300) provide, in response to linguistic information, efficient generation of a parametric representation of speech using a neural network. The method provides, in response to linguistic information efficient generation of a refined parametric representation of speech, comprising the steps of: A) using a data selection module to retrieve representative parameter vectors for each segment description according to the phonetic segment type and the phonetic segment types included in adjacent segment descriptions; B) interpolating between the representative parameter vectors according to the segment descriptions and duration to provide interpolated statistical parameters; C) converting the interpolated statistical parameters and linguistic information to neural network input parameters; D) utilizing a statistically enhanced neural network/neural network with post-processor to provide neural network output parameters that correspond to a parametric representation of speech; and converting the neural network output parameters to a refined parametric representation of speech.

Method, device and system for using statistical information to reduce

Text may be converted to audible signals, such as speech, by first training a neural network 106 using recorded audio messages 204. To begin the training, the recorded audio messages are converted into a series of audio frames 205 having a fixed duration 213. Then, each audio frame is assigned a phonetic representation 203 and a target acoustic representation 208, where the phonetic representation 203 is a binary word that represents the phone and articulation characteristics of the audio frame, while the target acoustic representation 208 is a vector of audio information such as pitch and energy. After training, the neural network 106 is used in conversion of text into speech. First, text that is to be convened is translated to a series of phonetic frames 401 of the same form as the phonetic representations 208 and having the fixed duration 213. Then the neural network produces acoustic representations in response to context descriptions 207 that include some of the phonetic frames 401. The acoustic representations are then converted into a speech wave form by a synthesizer 107.

Method and apparatus for converting text into audible signals using a

A totally self-checking memory cell array apparatus (30) has an array (31) of memory cells (32) selectively addressed by row and column decoders (33, 35) which receive unidirectional error detecting code signals as address inputs (34, 36). Data, as a multiple bit data word (A, B, C.sub.1, C.sub.2), is stored in the array (31) in unidirectional error detecting code form. Cells in each row (1-8) of the array have two separate row select connection lines (45 and 45a) for coupling the cell to data and data complement (46, 46*) connections. Error detection circuits (44, 47) are provided which determine errors by comparing the data and data complement lines for each data bit read out of the array and for detecting when multiple bit data words read out of the array are not coded in a unidirectional error detecting code format. The above apparatus provides error indications in case of any unidirectional errors in the row or column input address signals or the row or column decoders, or any unidirectional error corruption of the data stored in the memory cell array. This is achieved without completely duplicating each memory cell in the array and all row and column decoder circuitry.

Self-checking memory cell array apparatus

A method and means for maintaining continuous bit synchronization of data transmitted from a remote unit through a base unit to a landline unit is disclosed. An input data stream at a first bit rate, such as digitized or encrypted speech, is interleaved with a plurality of signalling words, and transmitted over an RF channel at a second bit rate. The base site recovers the clock of the received data, strips off the signalling word, modifies the bit rate of the received data, and adjusts the recovered clock rate to provide an output data stream which is in bit synchronization with the input data stream and within a predetermined modem specification. The encrypted data is then sent over landlines to a decryption unit which requires bit synchronization.

Growing community of inventors

Chicago, IL, United States of America

Gerald Edward Corrigan