Patent US6073096A successfully refreshed from Google Patents.
US6073096A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Speaker adaptation system and method based on class-specific pre-clustering training speakers
Assignee: International Business Machines Corp
•
Filing: 1998-02-04
•
Granted: 2000-06-06
•
Expires: 2018-02-04
Abstract
Abstract A method of speech recognition, in accordance with the present invention includes the steps of grouping acoustics to form classes based on acoustic features, clustering training speakers by the classes to provide class-specific cluster systems, selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a test speaker, transforming the subset of cluster systems to bring the subset of cluster systems closer to the test speaker based on the adaptation data to form adapted cluster systems and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the test speaker. System and methods for building speech recognition systems as well as adapting speaker systems for class-specific speaker clusters are included.
Inventors
- 👤 Yuqing Gao
- 👤 Mukund Padmanabhan
- 👤 Michael Alan Picheny
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 17:39
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/06; G10L15/065; G10L15/07
Claims Analysis
Total Claims: 49
Claim 1
Independent Claim
91 words
A method of speech recognition comprising the steps of: grouping acoustics to form classes based on acoustic features; clustering training speakers by the classes to provide class-specific cluster systems; selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 2
Independent Claim
91 words
A method of speech recognition comprising the steps of: grouping acoustics to form classes based on acoustic features; clustering training speakers by the classes to provide class-specific cluster systems; selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 3
Dependent Claim
24 words
The method of speech recognition as recited in claim 1, wherein the step of transforming is performed by employing a maximum a posteriori adaptation.
Claim 4
Dependent Claim
23 words
The method of speech recognition as recited in claim 1, wherein the step of transforming is performed by employing maximum likelihood linear regression.
Claim 5
Dependent Claim
55 words
The method of speech recognition as recited in claim 1, wherein the step of selecting further comprises: calculating a likelihood between the speaker and each cluster system; ranking each cluster system according to likelihood; and selecting a predetermined number of cluster systems closest to the speaker adaptation data to form the subset of cluster systems.
Claim 6
Dependent Claim
43 words
The method of speech recognition as recited in claim 1, further comprising the step of building class-specific cluster systems based on speaker dependent systems and partitions of acoustic space, each partition of acoustic space being characterized by a different set of acoustic features.
Claim 7
Dependent Claim
35 words
The method of speech recognition as recited in claim 1, wherein the step of transforming further comprises transforming the subset of cluster systems to adapt the subset of cluster systems with a speaker independent system.
Claim 8
Dependent Claim
20 words
The method of speech recognition as recited in claim 1, wherein each cluster system includes a Hidden Markov Model system.
Claim 9
Dependent Claim
22 words
The method of speech recognition as recited in claim 1, wherein the step of transforming is performed by employing a Bayesian adaptation.
Claim 10
Independent Claim
92 words
A method of building class-specific cluster systems comprising the steps of: providing a speaker dependent system for each of a plurality of training speakers; partitioning an acoustic space according to classes, each class being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces according to classes to build acoustic spaces with common features from all the speaker dependent systems; and clustering the grouped acoustic spaces with common features to form cluster systems based on acoustic characteristics of the speakers, the acoustic characteristics including class-specific characteristics.
Claim 11
Independent Claim
92 words
A method of building class-specific cluster systems comprising the steps of: providing a speaker dependent system for each of a plurality of training speakers; partitioning an acoustic space according to classes, each class being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces according to classes to build acoustic spaces with common features from all the speaker dependent systems; and clustering the grouped acoustic spaces with common features to form cluster systems based on acoustic characteristics of the speakers, the acoustic characteristics including class-specific characteristics.
Claim 12
Dependent Claim
31 words
The method of building class-specific cluster systems as recited in claim 9 further comprises the step of combining the cluster systems to form a cluster system for the full acoustic space.
Claim 13
Dependent Claim
26 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of clustering includes the step of clustering by a bottom-up method.
Claim 14
Dependent Claim
34 words
The method of building class-specific cluster systems as recited in claim 11, wherein the step of clustering by the bottom-up method includes measuring a distance between acoustic features between speakers by a Gaussian likelihood.
Claim 15
Dependent Claim
26 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of clustering includes the step of clustering by a top-down method.
Claim 16
Dependent Claim
34 words
The method of building class-specific cluster systems as recited in claim 13, wherein the step of clustering by the top-down method includes measuring a distance between acoustic features between speakers by a Euclidean distance.
Claim 17
Dependent Claim
35 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of clustering includes the step of clustering the grouped acoustic spaces to form cluster systems based on a common accent.
Claim 18
Dependent Claim
35 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of clustering includes the step of clustering the grouped acoustic spaces to form cluster systems based on a common gender.
Claim 19
Dependent Claim
62 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of providing a speaker dependent system for each of a plurality of training speakers further comprises the steps of: decoding training data of a speaker based on a speaker-independent system; and building the speaker dependent system by storing a set of labeled acoustic vectors for the speaker.
Claim 20
Dependent Claim
63 words
The method of building class-specific cluster systems as recited in claim 9, wherein the step of partitioning an acoustic space further includes the steps of: gathering expectation-maximization counts of a same context from a speaker independent system; generating a speaker independent context-independent system from the counts; and clustering the context-independent system to generate partitions of a whole acoustic space into the acoustic spaces.
Claim 21
Independent Claim
144 words
A method of speech recognition comprising the steps of: providing a speaker dependent system for each of a plurality of training speakers; providing an acoustic space for each of the training speakers, each acoustic space being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces to build acoustic spaces with common features from all the speaker dependent systems; clustering the grouped acoustic spaces to form cluster systems based on a common acoustic characteristic; selecting from a group of cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 22
Independent Claim
144 words
A method of speech recognition comprising the steps of: providing a speaker dependent system for each of a plurality of training speakers; providing an acoustic space for each of the training speakers, each acoustic space being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces to build acoustic spaces with common features from all the speaker dependent systems; clustering the grouped acoustic spaces to form cluster systems based on a common acoustic characteristic; selecting from a group of cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 23
Dependent Claim
52 words
The method of speech recognition as recited in claim 19, wherein the step of selecting further comprises: calculating a likelihood for each cluster system; ranking each cluster system according to likelihood; and selecting a predetermined number of cluster systems closest to the speaker adaptation data to form the subset of cluster systems.
Claim 24
Dependent Claim
52 words
The method of speech recognition as recited in claim 19, wherein the group of systems includes class-specific clusters and further comprises the step of building class-specific cluster systems based on speaker dependent systems and partitions of acoustic space, each partition of acoustic space being characterized by a different set of acoustic features.
Claim 25
Dependent Claim
26 words
The method of speech recognition as recited in claim 19, wherein each cluster system of the group of cluster systems includes a Hidden Markov Model system.
Claim 26
Dependent Claim
30 words
The method of speech recognition as recited in claim 19, further comprises the step of combining the cluster systems to form a speaker cluster system of all the acoustic spaces.
Claim 27
Dependent Claim
24 words
The method of speech recognition as recited in claim 19, wherein the step of clustering includes the step of clustering by a bottom-up method.
Claim 28
Dependent Claim
32 words
The method of speech recognition as recited in claim 24, wherein the step of clustering by the bottom-up method includes measuring a distance between acoustic features between speakers by a Gaussian likelihood.
Claim 29
Dependent Claim
24 words
The method of speech recognition as recited in claim 19, wherein the step of clustering includes the step of clustering by a top-down method.
Claim 30
Dependent Claim
32 words
The method of speech recognition as recited in claim 26, wherein the step of clustering by the top-down method includes measuring a distance between acoustic features between speakers by a Euclidean distance.
Claim 31
Dependent Claim
60 words
The method of speech recognition as recited in claim 19, wherein the step of providing a speaker dependent system for each of a plurality of training speakers further comprises the steps of: decoding training data of a speaker based on a speaker-independent system; and building the speaker dependent system by storing a set of labeled acoustic vectors for the speaker.
Claim 32
Dependent Claim
68 words
The method of speech recognition as recited in claim 19, wherein the step of providing an acoustic space for each of the training speakers further includes the steps of: gathering expectation maximization counts of a same context from a speaker independent system; generating a speaker independent context-independent system from the counts; and clustering the context-independent system to generate partition of a whole acoustic space into the acoustic spaces.
Claim 33
Independent Claim
98 words
A system for speech recognition comprising: means for grouping acoustics to form classes based on acoustic features; means for clustering training speakers by the classes to provide class-specific cluster systems; means for selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; means for transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and means for combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 34
Independent Claim
98 words
A system for speech recognition comprising: means for grouping acoustics to form classes based on acoustic features; means for clustering training speakers by the classes to provide class-specific cluster systems; means for selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; means for transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and means for combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 35
Dependent Claim
24 words
The system for speech recognition as recited in claim 30, wherein the means for transforming includes means for employing a maximum a posteriori adaptation.
Claim 36
Dependent Claim
23 words
The system for speech recognition as recited in claim 30, wherein the means for transforming includes means for employing maximum likelihood linear regression.
Claim 37
Dependent Claim
61 words
The system for speech recognition as recited in claim 30, wherein the means for selecting further comprises: means for calculating a likelihood between the speaker and each cluster system; means for ranking each cluster system according to likelihood; and means for selecting a predetermined number of cluster systems closest to the speaker adaptation data to form the subset of cluster systems.
Claim 38
Dependent Claim
51 words
The system for speech recognition as recited in claim 30, wherein the group of systems includes class-specific clusters and further comprising means for building class-specific cluster systems based on speaker dependent systems and partitions of acoustic space, each partition of acoustic space being characterized by a different set of acoustic features.
Claim 39
Dependent Claim
20 words
The system for speech recognition as recited in claim 30, wherein each cluster system includes a Hidden Markov Model system.
Claim 40
Independent Claim
149 words
A system for speech recognition comprising: a speaker dependent system for each of a plurality of training speakers; an acoustic space for each of the training speakers, each acoustic space being characterized by a set of acoustic features; means for grouping the speaker dependent systems with the acoustic spaces to build acoustic spaces with common features from all the speaker dependent systems; means for clustering the grouped acoustic spaces to form cluster systems based on a common acoustic characteristic; means for selecting from a group of cluster systems, a subset of cluster systems closest to adaptation data from a speaker; means for transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and means for combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 41
Independent Claim
149 words
A system for speech recognition comprising: a speaker dependent system for each of a plurality of training speakers; an acoustic space for each of the training speakers, each acoustic space being characterized by a set of acoustic features; means for grouping the speaker dependent systems with the acoustic spaces to build acoustic spaces with common features from all the speaker dependent systems; means for clustering the grouped acoustic spaces to form cluster systems based on a common acoustic characteristic; means for selecting from a group of cluster systems, a subset of cluster systems closest to adaptation data from a speaker; means for transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and means for combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 42
Dependent Claim
58 words
The system for speech recognition as recited in claim 36, wherein the means for selecting further comprises: means for calculating a likelihood for each cluster system; means for ranking each cluster system according to likelihood; and means for selecting a predetermined number of cluster systems closest to the speaker adaptation data to form the subset of cluster systems.
Claim 43
Dependent Claim
43 words
The system for speech recognition as recited in claim 36, wherein the group of cluster systems includes class-specific clusters based on speaker dependent systems and partitions of acoustic space, each partition of acoustic space being characterized by a different set of acoustic features.
Claim 44
Dependent Claim
25 words
The system for speech recognition as recited in claim 36, wherein each cluster system of the group of cluster systems includes a Hidden Markov Model.
Claim 45
Dependent Claim
24 words
The system for speech recognition as recited in claim 36, further comprises a speaker cluster system formed by combining all of the acoustic spaces.
Claim 46
Dependent Claim
110 words
A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for speech recognition, the method steps comprising: grouping acoustics to form classes based on acoustic features; clustering training speakers by the classes to provide class-specific cluster systems; selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 47
Dependent Claim
110 words
A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for speech recognition, the method steps comprising: grouping acoustics to form classes based on acoustic features; clustering training speakers by the classes to provide class-specific cluster systems; selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a speaker; transforming the subset of cluster systems to bring the subset of cluster systems closer to the speaker based on the adaptation data to form adapted cluster systems; and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the speaker.
Claim 48
Dependent Claim
115 words
A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for building class-specific cluster systems for speech recognition systems, the method steps comprising: providing a speaker dependent system for each of a plurality of training speakers; partitioning an acoustic space according to classes, each class being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces according to classes to build acoustic spaces with common features from all the speaker dependent systems; and clustering the grouped acoustic spaces with common features to form cluster systems based on acoustic characteristics of the speakers, the acoustic characteristics including class-specific characteristics.
Claim 49
Dependent Claim
115 words
A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for building class-specific cluster systems for speech recognition systems, the method steps comprising: providing a speaker dependent system for each of a plurality of training speakers; partitioning an acoustic space according to classes, each class being characterized by a set of acoustic features; grouping the speaker dependent systems with the acoustic spaces according to classes to build acoustic spaces with common features from all the speaker dependent systems; and clustering the grouped acoustic spaces with common features to form cluster systems based on acoustic characteristics of the speakers, the acoustic characteristics including class-specific characteristics.
References Cited (Backward Citations)
US5598507A
Cited
US5659662A
Cited
US5793891A
Cited
US5864810A
Cited
US5812975A
Cited
US5787394A
Cited
US5895447A
Cited
US5835890A
Cited
US5983178A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-06-06 | 2003-12-06 | 2004-06-06 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-06-06 | 2007-12-06 | 2008-06-06 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-06-06 | 2011-12-06 | 2012-06-06 | $7700.00 | Window Closed / Lapsed |