Patent
1991-12-20
1997-04-15
Hafiz, Tariq R.
395 24, 395 21, G10L 506, G10L 900
Patent
active
056218575
ABSTRACT:
Improved system and method for speaker-independent speech token recognition are described. The system is neural network-based and involves processing a sequence of spoken utterances, e.g. separately articulated letters of a name, to identify the same based upon a highest probability match of each utterance with learned speech tokens, e.g. the letters of the English language alphabet, and based upon a highest probability match of the uttered sequence with a defined utterance library, e.g. a list of names. First, the spoken utterance is digitized or captured and processed into a spectral representation. Second, discrete time frames of the DFT are classified phonetically. Third, the time-frame outputs are used by a modified Viterbi search to locate segment boundaries, near which such segment boundaries lies the information that is needed to discriminate letters. Fourth, the segmented or bounded representation is reclassified using such information into individual hypothesized letters. Fifth, successive, hypothesized letter scores are analyzed to obtain a high probability match with a spelled word within the utterance library. The system and method comprehend finer distinctions near points of interest used to discriminate difficult-to-recognize letter pair differences such as M/N, B/D, etc. The system is described in the context of phone line reception of names spelled by remote users.
REFERENCES:
patent: 4040215 (1977-08-01), Amano et al.
patent: 4752179 (1988-06-01), Levinson
patent: 4813076 (1989-03-01), Miller
patent: 4852170 (1989-07-01), Bordeaux
patent: 4852172 (1989-07-01), Taguchi
patent: 4856067 (1989-08-01), Yamada et al.
patent: 4905285 (1990-02-01), Allen et al.
patent: 4908865 (1990-03-01), Doddington et al.
patent: 4937872 (1990-06-01), Hopfield et al.
patent: 4944012 (1990-07-01), Morio et al.
patent: 4977599 (1990-12-01), Bahl et al.
patent: 5023912 (1991-06-01), Segawa
patent: 5121428 (1992-06-01), Uchiyama et al.
patent: 5212730 (1993-05-01), Wheatley et al.
patent: 5263097 (1993-11-01), Katz et al.
patent: 5278911 (1994-01-01), Bickerton
Mark Fanty and Ron Cole, "Speaker-Independent English Alphabet Recognition: Experiments with the E-Set", Proceedings of the International Conference on Spoken Language Processing, Kobe, Japan, Nov., 1990.
Richard P. Lippmann and Ben Gold, "Neural-Net Classifiers Useful for Speech Recognition", IEEE 1st Inter Conf. on Neural Networks Jun.21, 1987.
Mahesan Niranyan, Frank Fallside, "Speech Feature Extraction using Neural Networks". Lecture Notes in Computer Science, Feb. 15-17, 1990.
Mike Chong & Frank Fallside, "Classification & Regression Tree Neural Networks for Automatic Speech Recognition". Jul. 9-13, 1990. Inter Neural Net. Conf.
C. Rogers et al, "Neural Network Enhancement for a two Speaker Separation System".
Cole et al, "speaker independent vowel recognition: comparison of backpropagation and trained classification trees"; Proceedings of the twenty-third annual hawaii international conference on system sciences, pp. 132-141 vol. 1, 2-5 Jan. 1990.
Sawai et al, "TDNN-LR continuous speech recognition system using adaptive inremental TDNN training"; ICASSP '91, pp. 53-56, 1991.
Sawai et al, "Parallelism, hierarchy scaling in time-delay neural networks for spotting japanese phonemes/cv-syllables"; 1989 IEEE International conference on neural networks, pp. 11-81 to 11-88, 1989.
Cole et al, "Speaker-independent recognition of spoken english letters"; IJCNN, pp. 45-51 vol. 2, 17-21 Jun. 1990.
Rossen et al, "A connectionist model for consonant-vowel syllable recognition"; ICASSP 88, pp. 59-62 vol. 1, 11-14 Apr. 1988.
Creekmore et al, "A comparative study of five spectral representations for speaker-independent phonetic recognition"; Conference record of the twenty-fifth asilomar conference on signals, systems and computers, pp. 330-334 vol. 1, 4-6 Nov. 1991.
Gedo et al, "Automatic speaker recognition system using the discrete hartley transform and an artificial neural network"; Conference record of the 25rh Asilomar Conference on signals, systems and computers, pp. 1151-1154 vol. 2, 4-6 Nov. 1991.
Rogers et al, "Neural Network enhancement for a two speaker separation system"; ICASSP-89: 1989 International Conference on Acoustics, Speech and signal processing, pp. 357-360 vol. 1, 23-26 May 1989.
Leung et al, "Speech recognition using stochastic segment neural networks"; pp. 613-616 vol. 1, 23-26 Mar. 1992.
Cole Ronald A.
Fanty Mark A.
Hafiz Tariq R.
Oregon Graduate Institute of Science and Technology
LandOfFree
Method and system for identifying and recognizing speech does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for identifying and recognizing speech, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for identifying and recognizing speech will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-368303