Compilation of weighted finite-state transducers from decision trees

• US 5,806,032 A
• Filed: 06/14/1996
• Issued: 09/08/1998
• Est. Priority Date: 06/14/1996
• Status: Expired due to Term

• Current Assignee: AT&T Corporation (AT&T, Inc.), Nokia of America Corporation (Nokia Corporation), Sound View Innovations, LLC (Sound View Innovation Holdings, LLC)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Sproat, Richard William
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US08/663,767
• Time in Patent Office: 816 Days
• Field of Search: 395/2.64, 395/2.54, 395/2.66, 395/2.69, 395/2.44, 704/235, 704/245, 704/255, 704/257, 704/260, 704/9
• US Class Current: 704/255
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks

Abstract

A method for automatically converting a decision tree into one or more weighted finite-state transducers. Specifically, the method in accordance with an illustrative embodiment of the present invention processes one or more terminal (i.e., leaf) nodes of a given decision tree to generate one or more corresponding weighted rewrite rules. Then, these weighted rewrite rules are processed to generate weighted finite-state transducers corresponding to the one or more terminal nodes of the decision tree. In this manner, decision trees may be advantageously compiled into weighted finite-state transducers, and these transducers may then be used directly in various speech and natural language processing systems. The weighted rewrite rules employed herein comprise an extension of conventional rewrite rules, familiar to those skilled in the art.

Claims

1. An automated method for synthesizing speech sounds based on a finite-state representation of linguistic data generated based on one or more decision tree models of said linguistic data, each of the one or more decision tree models comprising one or more terminal nodes thereof, the method comprising the steps of:

generating one or more weighted rewrite rules based on one or more of the terminal nodes of the one or more decision tree models;

generating one or more weighted finite-state transducers based on one or more of the one or more weighted rewrite rules; and

synthesizing one or more of said speech sounds based on said one or more weighted finite-state transducers.

2. The method of claim 1 wherein the decision tree models comprise classification and regression trees.

3. The method of claim 2 wherein the decision tree models represent a phonetic realization of one or more phonemes by specifying a mapping of the phonemes into one or more allophones.

4. The method of claim 3 wherein one or more of the decision trees has been trained based on a linguistic database.

5. The method of claim 1 wherein the weighted rewrite rules comprise rules having a form φ

→

ψ

/λ

_-- ρ

, where φ

, λ

, and ρ

represent regular expressions, ψ

represents a weighted regular expression, and where the form φ

→

ψ

/λ

_-- ρ

represents that an occurrence of φ

in an input string is to be replaced by ψ

whenever the occurrence of φ

is preceded by λ and

succeeded by ρ

.

6. The method of claim 5 wherein the weighted regular expression represented by ψ

comprises a rational power series.

7. The method of claim 5 wherein the step of generating the weighted finite-state transducers comprises generating a weighted finite-state transducer for a corresponding weighted rewrite rule having the form φ

→

ψ

/λ

_-- ρ

by generating a composition of a first transducer, a second transducer, a third transducer, a fourth transducer and a fifth transducer, wherein;

the first transducer introduces a marker ">

" before every occurrence of ρ

;

the second transducer introduces markers "<

₁ " and "<

₂ " before each occurrence of φ

that is followed by a ">

" marker;

the third transducer replaces φ

with ψ

when φ

is preceded by a "<

₁ " marker and succeeded by a ">

" marker, and deletes each of the ">

" markers;

the fourth transducer admits only those input strings in which occurrences of a "<

₁ " marker are preceded by λ

, and deletes the "<

₁ " marker at each such occurrence; and

the fifth transducer admits only those input strings in which occurrences of a "<

₂ " marker are not preceded by λ

, and deletes the "<

₂ " marker at each such occurrence.

8. The method of claim 1 wherein a given one of the decision tree models represents an input symbol, S, the given decision tree model further comprising a root node and a plurality of branches labelled with corresponding regular expressions, and wherein the step of generating the weighted rewrite rules comprises:

selecting one of the terminal nodes of the given decision tree model;

determining a path from the root node of the given decision tree model to the selected terminal node thereof, the path comprising one or more of the branches of the given decision tree model;

intersecting one or more of the regular expressions which label the branches of the given decision tree model comprised in the determined path, generating a left context, LC, and a right context, RC, therefrom;

generating a weighted disjunction, D, of output symbols based on the selected terminal node; and

generating a weighted rewrite rule having a form S→

D/LC_-- RC, representing that an occurrence of the input symbol, S, in an input string is to be replaced by the weighted disjunction, D, whenever the occurrence of S is preceded by the left context, LC, and succeeded by the right context, RC.

9. The method of claim 1 wherein a given one of the decision tree models comprises a plurality of terminal nodes, wherein the step of generating the weighted rewrite rules comprises generating a plurality of weighted rewrite rules corresponding to the plurality of terminal nodes of the given decision tree model, and wherein the step of generating the weighted finite-state transducers comprises:

generating a plurality of weighted finite-state transducers corresponding to the plurality of generated weighted rewrite rules; and

intersecting the plurality of weighted finite-state transducers to produce a weighted finite-state transducer corresponding to the given decision tree model.

10. The method of claim 9 wherein the step of generating the weighted finite-state transducers further comprises intersecting the weighted finite-state transducer corresponding to the given decision tree model with one or more other weighted finite-state transducers corresponding to one or more other decision tree models to produce a weighted finite-state transducer corresponding to a plurality of decision tree models.

11. An apparatus for automatically generating a finite-state representation of linguistic data based on one or more decision tree models of said linguistic data, each of the one or more decision tree models comprising one or more terminal nodes thereof, the apparatus comprising:

means for generating one or more weighted rewrite rules based on one or more of the terminal nodes of the one or more decision tree models; and

means for generating one or more weighted finite-state transducers based on one or more of the one or more weighted rewrite rules.

12. The apparatus of claim 11 wherein the decision tree models comprise classification and regression trees.

13. The apparatus of claim 12 wherein the decision tree models represent a phonetic realization of one or more phonemes by specifying a mapping of the phonemes into one or more allophones.

14. The apparatus of claim 13 wherein one or more of the decision trees has been trained based on a linguistic database.

15. The apparatus of claim 11 wherein the weighted rewrite rules comprise rules having a form φ

→

ψ

/λ

_-- ρ

, where φ

, λ

, and ρ

represent regular expressions, ψ

represents a weighted regular expression, and where the form φ

→

ψ

/λ

_-- ρ

represents that an occurrence of φ

in an input string is to be replaced by ψ

whenever the occurrence of φ

is preceded by λ and

succeeded by ρ

.

16. The apparatus of claim 15 wherein the weighted regular expression represented by ψ

comprises a rational power series.

17. The apparatus of claim 15 wherein the means for generating the weighted finite-state transducers comprises means for generating a weighted finite-state transducer for a corresponding weighted rewrite rule having the form φ

→

ψ

/λ

_-- ρ

by generating a composition of a first transducer, a second transducer, a third transducer, a fourth transducer and a fifth transducer, wherein;

the first transducer introduces a marker ">

" before every occurrence of ρ

;

the second transducer introduces markers "<

₁ " and "<

₂ " before each occurrence of φ

that is followed by a ">

" marker;

the third transducer replaces φ

with ψ

when φ

is preceded by a "<

₁ " marker and succeeded by a ">

" marker, and deletes each of the ">

" markers;

the fourth transducer admits only those input strings in which occurrences of a "<

₁ " marker are preceded by λ

, and deletes the "<

₁ " marker at each such occurrence; and

the fifth transducer admits only those input strings in which occurrences of a "<

₂ " marker are not preceded by λ

, and deletes the "<

₂ " marker at each such occurrence.

18. The apparatus of claim 11 wherein a given one of the decision tree models represents an input symbol, S, the given decision tree model further comprising a root node and a plurality of branches labelled with corresponding regular expressions, and wherein the means for generating the weighted rewrite rules comprises:

means for selecting one of the terminal nodes of the given decision tree model;

means for determining a path from the root node of the given decision tree model to the selected terminal node thereof, the path comprising one or more of the branches of the given decision tree model;

means for intersecting one or more of the regular expressions which label the branches of the given decision tree model comprised in the determined path, generating a left context, LC, and a right context, RC, therefrom;

means for generating a weighted disjunction, D, of output symbols based on the selected terminal node; and

means for generating a weighted rewrite rule having a form S→

D/LC_-- RC, representing that an occurrence of the input symbol, S, in an input string is to be replaced by the weighted disjunction, D, whenever the occurrence of S is preceded by the left context, LC, and succeeded by the right context, RC.

19. The apparatus of claim 11 wherein a given one of the decision tree models comprises a plurality of terminal nodes, wherein the means for generating the weighted rewrite rules comprises means for generating a plurality of weighted rewrite rules corresponding to the plurality of terminal nodes of the given decision tree model, and wherein the means for generating the weighted finite-state transducers comprises:

means for generating a plurality of weighted finite-state transducers corresponding to the plurality of generated weighted rewrite rules; and

means for intersecting the plurality of weighted finite-state transducers to produce a weighted finite-state transducer corresponding to the given decision tree model.

20. The apparatus of claim 19 wherein the means for generating the weighted finite-state transducers further comprises means for intersecting the weighted finite-state transducer corresponding to the given decision tree model with one or more other weighted finite-state transducers corresponding to one or more other decision tree models to produce a weighted finite-state transducer corresponding to a plurality of decision tree models.

21. An automated method for recognizing speech sounds based on a finite-state representation of linguistic data generated based on one or more decision tree models of said linguistic data, each of the one or more decision tree models comprising one or more terminal nodes thereof, the method comprising the steps of:

generating one or more weighted rewrite rules based on one or more of the terminal nodes of the one or more decision tree models;

generating one or more weighted finite-state transducers based on one or more of

the one or more weighted rewrite rules; and

recognizing one or more of said speech sounds based on said one or more weighted finite-state transducers.

22. The method of claim 21 wherein the decision tree models comprise classification and regression trees.

23. The method of claim 22 wherein the decision tree models represent a phonetic realization of one or more phonemes by specifying a mapping of the phonemes into one or more allophones.

24. The method of claim 23 wherein one or more of the decision trees has been trained based on a linguistic database.

25. The method of claim 21 wherein the weighted rewrite rules comprise rules having a form φ

→

ψ

/λ

_-- ρ

, where φ

, λ

, and ρ

represent regular expressions, ψ

represents a weighted regular expression, and where the form φ

→

ψ

/λ

_-- ρ

represents that an occurrence of φ

in an input string is to be replaced by ψ

whenever the occurrence of φ

is preceded by λ and

succeeded by ρ

.

26. The method of claim 25 wherein the weighted regular expression represented by ψ

comprises a rational power series.

27. The method of claim 25 wherein the step of generating the weighted finite-state transducers comprises generating a weighted finite-state transducer for a corresponding weighted rewrite rule having the form φ

→

ψ

/λ

_-- ρ

by generating a composition of a first transducer, a second transducer, a third transducer, a fourth transducer and a fifth transducer, wherein;

the first transducer introduces a marker ">

" before every occurrence of ρ

;

the second transducer introduces markers "<

₁ " and "<

₂ " before each occurrence of φ

that is followed by a ">

" marker;

the third transducer replaces φ

with ψ

when φ

is preceded by a "<

₁ " marker and succeeded by a ">

" marker, and deletes each of the ">

" markers;

the fourth transducer admits only those input strings in which occurrences of a "<

₁ " marker are preceded by λ

, and deletes the "₁ <

" marker at each such occurrence; and

the fifth transducer admits only those input strings in which occurrences of a "<

₂ " marker are not preceded by λ

, and deletes the "<

₂ " marker at each such occurrence.

28. The method of claim 21 wherein a given one of the decision tree models represents an input symbol, S, the given decision tree model further comprising a root node and a plurality of branches labeled with corresponding regular expressions, and wherein the step of generating the weighted rewrite rules comprises:

selecting one of the terminal nodes of the given decision tree model;

determining a path from the root node of the given decision tree model to the selected terminal node thereof, the path comprising one or more of the branches of the given decision tree model;

intersecting one or more of the regular expressions which label the branches of the given decision tree model comprised in the determined path, generating a left context, LC, and a right context, RC, therefrom;

generating a weighted disjunction, D, of output symbols based on the selected terminal node; and

generating a weighted rewrite rule having a form S→

D/LC_-- RC, representing that an occurrence of the input symbol, S, in an input string is to be replaced by the weighted disjunction, D, whenever the occurrence of S is preceded by the left context, LC, and succeeded by the right context, RC.

29. The method of claim 21 wherein a given one of the decision tree models comprises a plurality of terminal nodes, wherein the step of generating the weighted rewrite rules comprises generating a plurality of weighted rewrite rules corresponding to the plurality of terminal nodes of the given decision tree model, and wherein the step of generating the weighted finite-state transducers comprises:

generating a plurality of weighted finite-state transducers corresponding to the plurality of generated weighted rewrite rules; and

intersecting the plurality of weighted finite-state transducers to produce a weighted finite-state transducer corresponding to the given decision tree model.

30. The method of claim 29 wherein the step of generating the weighted finite-state transducers further comprises intersecting the weighted finite-state transducer corresponding to the given decision tree model with one or more other weighted finite-state transducers corresponding to one or more other decision tree models to produce a weighted finite-state transducer corresponding to a plurality of decision tree models.