EP0387602A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Method and apparatus for the automatic determination of phonological rules as for a continuous speech recognition system
Assignee: International Business Machines Corp
•
Filing: 1990-02-27
•
Granted: —
•
Expires: 2010-02-27
Abstract
Abstract A continuous speech recognition system includes an automatic phonological rules generator which determines variations in the pronunciation of phonemes based on the context in which they occur. This phonological rules generator associates sequences of labels derived from vocalizations of a training text with respective phonemes inferred from the training text. These sequences are then annotated with their phoneme context from the training text and clustered into groups representing similar pronunciations of each phoneme. A decision tree is generated using the context information of the sequences to predict the clusters to which the sequences belong. The training data is processed by the decision tree to divide the sequences into leaf-groups representing similar pronunciations of each phoneme. The sequences in each leaf-group are clustered into sub-groups representing respectively different pronunciations of their corresponding phoneme in a give context. A Markov model is generated for each sub-group. The various Markov models of a leaf-group are combined into a single compound model by assigning common initial and final states to each model. The compound Markov models are used by a speech recognition system to analyze an unknown sequence of labels given its context.
Inventors
- 👤 Lalit Rai Bahl
- 👤 Peter Fitzhugh Brown
- 👤 Peter Vincent Desouza
- 👤 Robert Leroy Mercer
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-28 05:47
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/08; G10L15/14
Claims Analysis
Total Claims: 15
Claim 1
Independent Claim
133 words
A method for automatically generating phonological rules regarding the pronunciation of language components, comprising the steps of: A) processing a training text and vocalizations representing the training text to obtain a plurality of samples representing the language components of said vocalizations; B) selecting, from among the plurality of samples, a set of samples representing respective instances of a selected language component in the vocalizations; C) annotating each of said selected samples with an indicator of at least one language component in a contextual relationship with the said selected sample; D) generating, from said further annotated selected samples, a decision tree that separates the selected samples into respectively different leaf groups based on said context indicators, each of said leaf groups representing a pronunciation of said selected language component in a respectively different context.
Claim 2
Independent Claim
133 words
A method for automatically generating phonological rules regarding the pronunciation of language components, comprising the steps of: A) processing a training text and vocalizations representing the training text to obtain a plurality of samples representing the language components of said vocalizations; B) selecting, from among the plurality of samples, a set of samples representing respective instances of a selected language component in the vocalizations; C) annotating each of said selected samples with an indicator of at least one language component in a contextual relationship with the said selected sample; D) generating, from said further annotated selected samples, a decision tree that separates the selected samples into respectively different leaf groups based on said context indicators, each of said leaf groups representing a pronunciation of said selected language component in a respectively different context.
Claim 3
Dependent Claim
60 words
The method for automatically generating phonological rules set forth in Claim 1 wherein: Step C) further includes the steps of: C1) grouping said annotated selected samples into a plurality of clusters, each cluster representing a respectively different pronunciation of said selected language component; and C2) further annotating each selected sample with an indicator of the cluster to which it belongs;
Claim 4
Dependent Claim
127 words
The method for automatically generating phonological rules set forth in Claim 2 wherein: said language components are phonemes; said samples are sequences of fenemes having distinct types, each of said sequences of fenemes corresponding to a respective one of said phonemes; and step C1) includes the steps of: assigning each sequence of fenemes to a respectively different prototype cluster; calculating the expected frequency for each type of feneme in the prototype cluster; statistically comparing the expected frequency for the fenemes in each prototype cluster to the expected feneme frequencies for all other prototype clusters to generate a statistical difference value for each pair of clusters; combining pairs of prototype clusters that exhibit a statistical difference which is less than a threshold value to generate new prototype clusters.
Claim 5
Dependent Claim
146 words
The method for automatically generating phonological rules set forth in Claim 3 wherein the step of statistically comparing each prototype cluster to all other prototype clusters includes the steps of: generating a plurality of probabilistic models, each model representing a Markov model of the pronunciation of the phoneme represented by a respectively different one of said prototype clusters; generating a plurality of histograms, each histogram representing the relative frequency of occurrence of each feneme in a respectively different one of said prototype clusters; calculating, for each pair of prototype clusters, a log-likelihood ratio that the histograms for each prototype clusters in said pair match the expected feneme frequencies of a single probabilistic model representing a combination of the respective probabilistic models for said pair of prototype clusters, wherein a sign inverted version of said log-likelihood ratio is the statistical difference value for said pair of clusters.
Claim 6
Dependent Claim
44 words
The method for automatically generating phonological rules set forth in Claim 1, further comprising the step of generating, for each leaf group of said decision tree, a probabilistic model representing the pronunciation of the language component represented by the samples of said leaf group.
Claim 7
Dependent Claim
34 words
The method for automatically generating phonological rules set forth in Claim 5, wherein said probabilistic model has the form of a Markov model representing the relative frequencies of the samples of the leaf group.
Claim 8
Dependent Claim
110 words
The method for automatically generating phonological rules set forth in Claim 5 wherein the step of generating a probabilistic model for each leaf of the group includes the steps of: grouping the samples of the leaf into a plurality of clusters, each cluster representing a respectively different pronunciation of the language component represented by the samples of the leaf group; generating, from said plurality of clusters, a respective plurality of statistical models each of said statistical models having the form of a Markov model; and augmenting said plurality of statistical models by adding a common initial state and a common final state to each model to generate said probabilistic model.
Claim 9
Independent Claim
142 words
A method for automatically generating a decision graph relating to a set of sample values comprising the steps of: A) annotating each of the sample values in said set with an indicator of a first attribute of said sample values; B) further annotating each of the sample values in said set with an indicator of a second attribute of said sample values, said set of further annotated samples having a predetermined entropy value measured with respect to said second attribute indicator; and C) generating a test question, in terms of the first attribute indicator of said further annotated samples, which tends to divide said further annotated sample values into first and second groups, wherein said first and second groups of samples have a combined entropy value, measured with respect to said second attribute indicator, that is less than said predetermined entropy value.
Claim 10
Independent Claim
142 words
A method for automatically generating a decision graph relating to a set of sample values comprising the steps of: A) annotating each of the sample values in said set with an indicator of a first attribute of said sample values; B) further annotating each of the sample values in said set with an indicator of a second attribute of said sample values, said set of further annotated samples having a predetermined entropy value measured with respect to said second attribute indicator; and C) generating a test question, in terms of the first attribute indicator of said further annotated samples, which tends to divide said further annotated sample values into first and second groups, wherein said first and second groups of samples have a combined entropy value, measured with respect to said second attribute indicator, that is less than said predetermined entropy value.
Claim 11
Dependent Claim
65 words
The method for automatically generating a decision graph set forth in Claim 9 wherein step C) includes the steps of: grouping the annotated sample values in said set into a plurality of clusters according to the second attribute of said sample values; and further annotating each of said annotated samples with an indicator of the cluster to which it belongs as said second attribute indicator.
Claim 12
Dependent Claim
135 words
The method for automatically generating a decision graph set forth in Claim 10, further including the steps of: generating a first further test question in terms of the first attribute indicator of the sample values in said second group which transfers sample values from said second group to said first group to form revised first and second groups of sample values having a reduced combined entropy value measured with respect to said second attribute indicator; and generating a second further test question in terms of the first attribute indicator of said sample values in said first group which transfers sample values from said first group to said second group to form further revised first and second groups of sample values having a further reduced combined entropy value measured with respect to said second attribute indicator.
Claim 13
Dependent Claim
64 words
The method for automatically generating a decision graph set forth in Claim 11, wherein: said sample values are respective sequences of fenemes representing respective vocalizations of phonemes; said first attribute of said sample values relates to the phonetic contexts in which the sequences of fenemes occur; and said second attribute of said sample values relates to the vocalizations represented by said sequences of fenemes.
Claim 14
Dependent Claim
126 words
The method for automatically generating a decision graph set forth in Claim 10, wherein each sample values in said set of sample values includes a sequence of subsample values having distinct types and the step of grouping the annotated sample values in said set into clusters includes the steps of: assigning each sample value to a respectively different prototype cluster; calculating an expected frequency for each type of subsample in each prototype cluster; statistically comparing the expected subsample frequency of each prototype cluster to the expected subsample frequencies of all other prototype clusters to generate a statistical difference value for each pair of clusters; combining pairs of prototype clusters that exhibit a statistical difference which is less than a threshold value to generate new prototype clusters.
Claim 15
Dependent Claim
158 words
The method for automatically generating a decision graph set forth in Claim 13 wherein the step of statistically comparing the expected subsample frequency of each prototype cluster to the expected subsample frequencies of all other prototype clusters includes the steps of: generating a plurality of probabilistic models, each model representing a Markov model of the sequences of subsamples represented by a respectively different one of said prototype clusters; generating a plurality of histograms, each histogram representing the relative frequency of occurrence of each type of subsample in a respectively different one of said prototype clusters; calculating, for each pair of prototype clusters, a log-likelihood ratio that the histograms for each prototype clusters in said pair match the expected subsample frequencies of a single probabilistic model representing a combination of the respective probabilistic models for said pair of prototype clusters, wherein a sign inverted version of said log-likelihood ratio is the statistical difference value for said pair of clusters.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.