Operator interactions for developing phoneme recognition by neur

Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

704231, 704259, G10L 506, G10L 900

Patent

active

058678166

ABSTRACT:
An automated speech recognition system converts a speech signal into a compact, coded representation that correlates to a speech phoneme set. A number of different neural network pattern matching schemes may be used to perform the necessary speech coding. An integrated user interface guides a user unfamiliar with the details of speech recognition or neural networks to quickly develop and test a neural network for phoneme recognition. To train the neural network, digitized voice data containing known phonemes that the user wants the neural network to ultimately recognize are processed by the integrated user interface. The digitized speech is segmented into phonemes with each segment being labelled with a corresponding phoneme code. Based on a user selected transformation method and transformation parameters, each segment is transformed into a series of multiple dimension vectors representative of the speech characteristics of that segment. These vectors are iteratively presented to a neural network to train/adapt that neural network to consistently distinguish and recognize these vectors and assign an appropriate phoneme code to each vector. Simultaneous display of the digitized speech, segments, vector sets, and a representation of the trained neural network assist the user in visually confirming the acceptability of the phoneme training set. A user may also selectively audibly confirm the acceptability of the digitization scheme, the segments, and the transform vectors so that satisfactory training data are presented to the neural network. If the user finds a particular step or parameter produces an unacceptable result, the user may modify one or more of the parameters and verify whether the modification effected an improvement in performance. The trained neural network is also automatically tested by presenting a test speech signal to the integrated user interface and observing both audibly and visually automatic segmentation of the speech, transformation into multidimensional vectors, and the resulting neural network assigned phoneme codes. A method of decoding such phoneme codes using the neural network is also disclosed.

REFERENCES:
patent: 5040214 (1991-08-01), Grossberg et al.
patent: 5163111 (1992-11-01), Baji et al.
patent: 5317673 (1994-05-01), Cohen
patent: 5426745 (1995-06-01), Baji
Dalsgaard P. et al., Interactive Labelling of Multi-Lingual Speech Corpora, ICASSP 92, pp. I-549-52.
Hwang J-N, Hang, OL., Interactive Query Learning for Isolated Pattern Recognition, NNSP II '92, pp. 39-94.
Handbook of Neural Computing Applications, by A. Marren et al., 1990, pp. 92-103, 142-145, 402-403.
Data Translation, Inc. 1993 Applications Handbook, vol. 2, No. 1, "For the Unwary, Aliasing Generates Problems in Unexpected Areas," Jul. 1992, by R. Calkins, p. 125.
Creative Labs, Inc. Sound Blaster Pro User Reference Manual, Feb. 1992, pp. 11-12.
Signal Analysis, by Athoansio Papoulis, 1977, Chapter 3-4 "Discrete Fourier Series and Fast Fourier Transforms," p. 79.
Neuralware, Inc. Neural Computing,"Back-Propagation," 1991, pp. NC-89-NC94.
Neuralware, Inc. Reference Guide, 1991, pp. RF-20, RF-36-RF-40, RF-171-RF-214, RF-217-220.
Technology--The Freedom to Dial Hands-Free, "Voice Recognition Technology Provides Hand's Free Dialing for Cellular Phone Users," by S. Bayle, Mar. 1993, pp. 36-38.
NASA Tech Briefs, "Voice-Recognition System Records Inspection Data," by L. Rochester, Jan. 1993, p. 50.
IEEE, "Application of Fullly Recurrent Neural Networks for Speech Recognition," by S. Lee et al., 1991, pp. 77, 79.
Proceedings of the IEEE, vol. 78, No. 10, "Neural Network Models of Sensory Integration for Improved Vowel Recognition," by B. Yuhas et al., Oct. 1990, pp. 1658, 1660.
Neural Networks, vol. 3, "Speaker-Independent Isolated Digit Recognition: Multilayer Perceptrons vs. Dynamic Time Warping," by L. Bottou et al., 1990, pp. 453-455.
Neural Computation, "Review of Neural Networks for Speech Recognition," by R. Lipmann, 1989, pp.1-37.
IEEE, "Integrated Training for Spotting Japanese Phonemes Using Large Phonemic Time-Delay Neural Networks," by M. Miyatake et al., 1990, pp. 449-450.
Numerical Recipes in C--The Art of Scientific Computing, by W. Press et al., 1990, pp. 447-466.
"A Novel Objective Function for Improved Phoneme Recognition Using Time Delay Neural Networks," by J.B. Hampshire II et al., pp. I-235-I-241.
Automated Speech Recognition, by K. Lee, pp. 43-50.
"On the Use of Neural Networks and Fuzzy Logic in Speech Recognition," by A. Amano et al., p. I-301.
IEEE Transactions on Neural Networks, vol. 1, No. 2, "Parallel, Self-Organizing, Hierarchical Neural Networks," by O. Ersoy et al., Jun. 1990, pp. 167-168.
IEEE Transactions on Neural Networks, vol. 2, No. 1, "Differential Competitive Learning for Centroid Estimation and Phoneme Recognition," by S. Kong et al., Jan. 1991, pp. 118, 120, 123.
"A Spatio-Temporal Pattern Recognition Approach to Word Recognition," by M. Tom et al., Fev. 1989, p. I-351.
Speaker-Independent Phoneme Recognition on TIMIT Database Using Integrated Time-Delay Neural Networks (TDNNs), by N. Hataoka et all., 1989, pp. I-57-I-59.
Neural Computation, "Modular Construction of Time-delay Neural Networks for Speech Recognition," By A. Lippmann, 1989, pp. 1, 2, 41.
"Representional Issues in a Neural Network Model of Syllable Recognition," by M. Rossen et al., 1989, pp. I-19, I-21.
Neural Networks, vol. 3, "A Time-Delay Neural Network Architecture for Isolated Word Recognition," by K. Lang et al., 1990, pp. 23, 28.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Operator interactions for developing phoneme recognition by neur does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Operator interactions for developing phoneme recognition by neur, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Operator interactions for developing phoneme recognition by neur will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-1127219

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.