US6182039B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus using probabilistic language model based on confusable sets for speech recognition
Assignee: Panasonic Holdings Corp
•
Filing: 1998-03-24
•
Granted: 2001-01-30
•
Expires: 2018-03-24
Abstract
Abstract The speech recognizer incorporates a language model that reduces the number of acoustic pattern matching sequences that must be performed by the recognizer. The language model is based on knowledge of a pre-defined set of syntactically defined content and includes a data structure that organizes the content according to acoustic confusability. A spelled name recognition system based on the recognizer employs a language model based on classes of letters that the recognizer frequently confuses for one another. The language model data structure is optionally an N-gram data structure, a tree data structure, or an incrementally configured network that is built during a training sequence. The incrementally configured network has nodes that are selected based on acoustic distance from a predetermined lexicon.
Inventors
- 👤 Luca Rigazio
- 👤 Jean-claude Junqua
- 👤 Michael Galler
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 11:36
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/08; G10L15/18; G10L15/183; G10L15/19; G10L15/193; G10L15/197
Claims Analysis
Total Claims: 18
Claim 1
Independent Claim
128 words
A speech recognition system for analyzing input speech corresponding to a predefined set of syntactically defined content comprising: a speech recognizer that performs a recognition process on input speech by considering a plurality of acoustic pattern matching sequences; and a language model associated with said recognizer for constraining the number of sequences considered by the recognizer during said recognition process; wherein said language model is based on knowledge of said predefined set of syntactically defined content and includes a data structure that organizes said syntactically defined content according to acoustic confusability; wherein said predefined set of syntactically defined content is a set of sound units comprised of a collection of sub-units and wherein said data structure defines a plurality of confusable sets to which said sub-units are assigned.
Claim 2
Dependent Claim
128 words
A speech recognition system for analyzing input speech corresponding to a predefined set of syntactically defined content comprising: a speech recognizer that performs a recognition process on input speech by considering a plurality of acoustic pattern matching sequences; and a language model associated with said recognizer for constraining the number of sequences considered by the recognizer during said recognition process; wherein said language model is based on knowledge of said predefined set of syntactically defined content and includes a data structure that organizes said syntactically defined content according to acoustic confusability; wherein said predefined set of syntactically defined content is a set of sound units comprised of a collection of sub-units and wherein said data structure defines a plurality of confusable sets to which said sub-units are assigned.
Claim 3
Dependent Claim
19 words
The system of claim 1 wherein said acoustic confusability is established based on recognition performance of said speech recognizer.
Claim 4
Dependent Claim
14 words
The system of claim 1 wherein said data structure is a bigram data structure.
Claim 5
Dependent Claim
22 words
The system of claim 1 wherein said data structure is an N-gram data structure, where N is an integer greater than one.
Claim 6
Dependent Claim
14 words
The system of claim 1 wherein said data structure is a tree data structure.
Claim 7
Dependent Claim
25 words
The system of claim 1 wherein said data structure is an incrementally configured network having nodes selected based on acoustic distance from a predetermined lexicon.
Claim 8
Dependent Claim
19 words
The recognition system of claim 1 wherein the confusable sets are defined by sub-units that the recognizer finds confusable.
Claim 9
Dependent Claim
69 words
The recognition system of claim 1 wherein the sound units include spelled words and the sub-units include individual letters, the confusable sets including at least one of the following sets and consisting essentially of the following letters: A-Set: a, h, j, k E-Set: b, c, d, e, g, p, t, v, z F-Set: f, s, x I-Set: i, y, r M-Set: l, m, n, o U-Set: u, q, w.
Claim 10
Dependent Claim
21 words
The recognition system of claim 1 wherein the sound units include phonetic transcriptions of words and the sub-units include individual phonemes.
Claim 11
Independent Claim
186 words
A method for analyzing input speech with a speech recognizer of the type that performs a recognition process on input speech by considering a plurality of acoustic pattern matching sequence, comprising the steps of: constructing an individual language model bas ed on confusable classes; and using said language model to constrain the number of sequences considered by said speech recognizer during the recognition process; wherein said language model is based on knowledge of a predefined set of syntactically defined content and includes a data structure that organizes said syntactically defined content according to acoustic confusability; wherein said predefined set of syntactically defined content is a set of spelled words or names comprised of collection of individual letters and wherein said data structure defines a plurality of confusable classes to which said individual letters are assigned, wherein said confusable sets include at least one of the following sets and consisting essentially of the following letters: A-Set: a, h, j, k E-Set: b, c, d, e, g, p, t, v, z F-Set: f, s, x I-Set: I, y, r M-Set: l, m, n, o U-Set: u, q, w.
Claim 12
Independent Claim
186 words
A method for analyzing input speech with a speech recognizer of the type that performs a recognition process on input speech by considering a plurality of acoustic pattern matching sequence, comprising the steps of: constructing an individual language model bas ed on confusable classes; and using said language model to constrain the number of sequences considered by said speech recognizer during the recognition process; wherein said language model is based on knowledge of a predefined set of syntactically defined content and includes a data structure that organizes said syntactically defined content according to acoustic confusability; wherein said predefined set of syntactically defined content is a set of spelled words or names comprised of collection of individual letters and wherein said data structure defines a plurality of confusable classes to which said individual letters are assigned, wherein said confusable sets include at least one of the following sets and consisting essentially of the following letters: A-Set: a, h, j, k E-Set: b, c, d, e, g, p, t, v, z F-Set: f, s, x I-Set: I, y, r M-Set: l, m, n, o U-Set: u, q, w.
Claim 13
Dependent Claim
24 words
The method of claim 2 wherein said step of constructing a language model is performed by grouping syntactically defined content according to acoustic confusability.
Claim 14
Dependent Claim
18 words
The method of claim 11 wherein acoustic confusability is established based on recognition performance of said speech recognizer.
Claim 15
Dependent Claim
16 words
The method of claim 10 further comprising organizing said language model into a bigram data structure.
Claim 16
Dependent Claim
24 words
The method of claim 10 further comprising organizing said language model into an N-gram data structure, where N is an integer greater than one.
Claim 17
Dependent Claim
16 words
The method of claim 10 further comprising organizing said language model into a tree data structure.
Claim 18
Dependent Claim
32 words
The method of claim 10 wherein the step of constructing a language model is performed by constructing an incrementally configured network having nodes selected based on acoustic distance from a predetermined lexicon.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2004-01-30 | 2004-07-31 | 2005-01-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2008-01-30 | 2008-07-31 | 2009-01-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2012-01-30 | 2012-07-31 | 2013-01-30 | $7700.00 | Window Closed / Lapsed |