Abstract: An improved method of providing out-of-vocabulary word rejection is achieved by adapting an initial garbage model (23) using received incoming enrollment speech (21).

Abstract: A network/portable computer (18) includes a modular bay (20) which can receive either a hard drive (22) or a data transceiver (24) for establishing a wireless network connection. When the network/portable computer (18) is in the local area covered by the network (10), the data transceiver (24) is installed to use the network (10) for storage and retrieval of data. When used in a remote area, where a network connection cannot be established, the hard drive (22) is installed in the modular bay (20) to provide a full featured computer.

Abstract: This invention consists of three enhancements to HMM-based automatic language identification systems. The three enhancements are: (i) language-discriminant acoustic model training and recognition, (ii) an acoustic model pruning procedure that picks only those phonetic models which are considered useful for language identification, and (iii) a neural network-based language classification method that uses knowledge-based features derived from phone sequences output by the HMM phonetic recognizers.

Abstract: The present invention is a speech processing system and method which includes a codebook generator (26) for generating a reference vector quantization codebook which describes a reference environment and for generating at least one secondary vector quantization codebook which describes at least one secondary environment. The secondary vector quantization codebooks are generated using the reference vector quantization codebook. A speech recognizer (38), which is trained using the reference vector quantization codebook, is also included. A pre-processor (32) accepts as input speech collected from an unknown environment and pre-processes that speech, before input to the speech recognizer (38), using the reference vector quantization codebook and an adaptation of the secondary vector quantization codebooks.

Abstract: Method of encoding speech signals which is based upon determining the roots of the linear prediction polynomial describing the spectrum of an analog speech signal, wherein the roots are candidates in determining the formants of the speech signal. The method involves the analysis of respective frames of sampled digital speech data using a linear predictive technique to determine a set of reflection coefficients or K-parameters which are then converted into the equivalent predictor coefficients or A-parameters describing a prediction polynomial having a plurality of roots corresponding to the poles of an all-pole filter characterizing the vocal tract. A modified Bairstow technique is then empolyed for factoring out quadratic factors which are then sorted in an ordered arrangement in terms of ascending bandwidths.

Abstract: Speaker-independent word recognition method and system for identifying individual spoken words based upon an acoustically distinct vocabulary of a limited number of words. The word recognition system may employ memory storage associated with a microprocessor or microcomputer in which reference templates of digital speech data representative of a limited number of words comprising the word vocabulary are stored. The word recognition system accepts an input analog speech signal from a microphone as derived from a single word-voice command spoken by any speaker. The analog speech signal is directed to an energy measuring circuit and a zero-crossing detector for determining a sequence of feature vectors based upon the zero-crossing rate and energy measurements of the sampled analog speech signal.

Abstract: Speaker-independent word recognition is performed, based on a small acoustically distinct vocabulary, with minimal hardware requirements. After a simple preconditioning filter, the zero crossing intervals of the input speech are measured and sorted by duration, to provide a rough measure of the frequency distribution within each input frame. The distribution of zero crossing intervals is transformed into a binary feature vector, which is compared with each reference template using a modified Hamming distance measure. A dynamic time warping algorithm is used to permit recognition of various speaker rate, and to economize on the reference template storage requirements. A mask vector for each reference template is used to ignore insignificant (or speaker-dependent) features of the words detected.

Abstract: Speaker-independent word recognition is performed, based on a small acoustically distinct vocabulary, with minimal hardware requirements. After a simple preconditioning filter, the zero crossing intervals of the input speech are measured and sorted by duration, to provide a rough measure of the frequency distribution within each input frame. The distribution of zero crossing intervals is transformed into a binary feature vector, which is compared with each reference template using a modified Hamming distance measure. A dynamic time warping algorithm is used to permit recognition of various speaker rates, and to economize on the reference template storage requirements. A mask vector with each reference vector on a template is used to ignore insignificant (or speaker-dependent) features of the words detected.