Discriminative training of document transcription system
First Claim
1. A method for use with a system including a first document containing at least some information in common with a spoken audio stream, the method performed by at least one computer processor executing computer program instructions to perform steps of:
- (A) identifying text in the first document, wherein the text represents a concept;
(B) identifying, based on the identified text and a repository of finite state grammars, a plurality of spoken forms of the concept, including at least one spoken form not contained in the first document, wherein all of the plurality of spoken forms have the same content as each other;
(C) replacing the identified text with a finite state grammar specifying the plurality of spoken forms of the concept to produce a second document;
(D) generating a document-specific language model based on the second document, comprising generating at least some of the document-specific language model based on the finite state grammar;
(E) using the document-specific language model in a speech recognition process to recognize the spoken audio stream and thereby to produce a third document;
(F) filtering text from the third document by reference to the second document to produce a filtered document in which text filtered from the third document is marked as unreliable; and
(G) using the filtered document and the spoken audio stream to train an acoustic model by performing steps of;
(G)(1) applying a first speech recognition process to the spoken audio stream using a set of base acoustic models and a grammar network based on the filtered document to produce a first set of recognition structures;
(G)(2) applying a second speech recognition process to the spoken audio stream using the set of base acoustic models and a second language model to produce a second set of recognition structures; and
(G)(3) performing discriminative training of the acoustic model using the first set of recognition structures, the second set of recognition structures, the filtered document, and only those portions of the spoken audio stream corresponding to text not marked as unreliable in the filtered document.
11 Assignments
0 Petitions
Accused Products
Abstract
A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model using discriminative training techniques, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.
14 Citations
8 Claims
-
1. A method for use with a system including a first document containing at least some information in common with a spoken audio stream, the method performed by at least one computer processor executing computer program instructions to perform steps of:
-
(A) identifying text in the first document, wherein the text represents a concept; (B) identifying, based on the identified text and a repository of finite state grammars, a plurality of spoken forms of the concept, including at least one spoken form not contained in the first document, wherein all of the plurality of spoken forms have the same content as each other; (C) replacing the identified text with a finite state grammar specifying the plurality of spoken forms of the concept to produce a second document; (D) generating a document-specific language model based on the second document, comprising generating at least some of the document-specific language model based on the finite state grammar; (E) using the document-specific language model in a speech recognition process to recognize the spoken audio stream and thereby to produce a third document; (F) filtering text from the third document by reference to the second document to produce a filtered document in which text filtered from the third document is marked as unreliable; and (G) using the filtered document and the spoken audio stream to train an acoustic model by performing steps of; (G)(1) applying a first speech recognition process to the spoken audio stream using a set of base acoustic models and a grammar network based on the filtered document to produce a first set of recognition structures; (G)(2) applying a second speech recognition process to the spoken audio stream using the set of base acoustic models and a second language model to produce a second set of recognition structures; and (G)(3) performing discriminative training of the acoustic model using the first set of recognition structures, the second set of recognition structures, the filtered document, and only those portions of the spoken audio stream corresponding to text not marked as unreliable in the filtered document. - View Dependent Claims (2, 3, 4)
-
-
5. A non-transitory computer-readable medium comprising computer program instructions executable by at least one computer processor to perform a method for use with a system, the system including a first document containing at least some information in common with a spoken audio stream, the method comprising:
-
(A) identifying text in the first document, wherein the text represents a concept; (B) identifying, based on the identified text and a repository of finite state grammars, a plurality of spoken forms of the concept, including at least one spoken form not contained in the first document, wherein all of the plurality of spoken forms have the same content as each other; (C) replacing the identified text with a finite state grammar specifying the plurality of spoken forms of the concept to produce a second document; (D) generating a document-specific language model based on the second document, comprising generating at least some of the document-specific language model based on the finite state grammar; (E) using the document-specific language model in a speech recognition process to recognize the spoken audio stream and thereby to produce a third document; (F) filtering text from the third document by reference to the second document to produce a filtered document in which text filtered from the third document is marked as unreliable; and (G) using the filtered document and the spoken audio stream to train an acoustic model by performing steps of;
comprising;(G)(1) applying a first speech recognition process to the spoken audio stream using a set of base acoustic models and a grammar network based on the filtered document to produce a first set of recognition structures; (G)(2) applying a second speech recognition process to the spoken audio stream using the set of base acoustic models and a second language model to produce a second set of recognition structures; and (G)(3) means for performing discriminative training of the acoustic model using the first set of recognition structures, the second set of recognition structures, the filtered document, and only those portions of the spoken audio stream corresponding to text not marked as unreliable in the filtered document. - View Dependent Claims (6, 7, 8)
-
Specification