Input generation for classifier
First Claim
Patent Images
1. A computer-implemented method for improving performance of a speech recognition system, comprising:
- generating a single text data structure for a classifier of a speech recognition system, including;
obtaining first n-best hypotheses as an output of a speech recognition task performed by automatic speech recognition (ASR) for an utterance received by the speech recognition system; and
combining the first n-best hypotheses horizontally in a predetermined order with a separator between each pair of n-best hypotheses to generate the single text data structure, wherein each separator is set based on a classification algorithm of the classifier of the speech recognition system as a symbol that is usable by the classifier of the speech recognition system; and
outputting the single text data structure as an input to the classifier to perform a classification task;
wherein the classifier is trained to perform the classification task based on a single training text data structure by;
obtaining source training data including a plurality of second n-best hypotheses and a transcription for each utterance from a database;
arranging the source training data with the transcription at a head or at an end of the second n-best hypotheses depending on the predetermined order to generate the single training text data structure; and
outputting the single training text data structure to the classifier.
1 Assignment
0 Petitions
Accused Products
Abstract
A computer-implemented method for generating an input for a classifier. The method includes obtaining n-best hypotheses which is an output of an automatic speech recognition (ASR) for an utterance, combining the n-best hypotheses horizontally in a predetermined order with a separator between each pair of hypotheses, and outputting the combined n-best hypotheses as a single text input to a classifier.
18 Citations
9 Claims
-
1. A computer-implemented method for improving performance of a speech recognition system, comprising:
-
generating a single text data structure for a classifier of a speech recognition system, including; obtaining first n-best hypotheses as an output of a speech recognition task performed by automatic speech recognition (ASR) for an utterance received by the speech recognition system; and combining the first n-best hypotheses horizontally in a predetermined order with a separator between each pair of n-best hypotheses to generate the single text data structure, wherein each separator is set based on a classification algorithm of the classifier of the speech recognition system as a symbol that is usable by the classifier of the speech recognition system; and outputting the single text data structure as an input to the classifier to perform a classification task; wherein the classifier is trained to perform the classification task based on a single training text data structure by; obtaining source training data including a plurality of second n-best hypotheses and a transcription for each utterance from a database; arranging the source training data with the transcription at a head or at an end of the second n-best hypotheses depending on the predetermined order to generate the single training text data structure; and outputting the single training text data structure to the classifier. - View Dependent Claims (2, 3)
-
-
4. An apparatus comprising:
-
a processor; and one or more non-transitory computer readable storage mediums collectively including instructions that, when executed by the processor, cause the processor to perform a method for improving performance of a speech recognition system, the method comprising; generating a single text data structure for a classifier of a speech recognition system, including; obtaining first n-best hypotheses as an output of a speech recognition task performed by automatic speech recognition (ASR) for an utterance received by the speech recognition system; and combining the first n-best hypotheses horizontally in a predetermined order with a separator between each pair of n-best hypotheses to generate the single text data structure, wherein each separator is set based on a classification algorithm of the classifier of the speech recognition system as a symbol that is usable by the classifier of the speech recognition system; and outputting the single text data structure as an input to the classifier to perform a classification task; wherein the classifier is trained to perform the classification task based on a single training text data structure by; obtaining source training data including a plurality of second n-best hypotheses and a transcription for each utterance from a database; arranging the source training data with the transcription at a head or at an end of the second n-best hypotheses depending on the predetermined order to generate the single training text data structure; and outputting the single training text data structure to the classifier. - View Dependent Claims (5, 6)
-
-
7. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a computer to cause the computer to perform a method for improving performance of a speech recognition system, the method comprising:
-
generating a single text data structure for a classifier of a speech recognition system, including; obtaining first n-best hypotheses as an output of a speech recognition task performed by automatic speech recognition (ASR) for an utterance received by the speech recognition system; and combining the first n-best hypotheses horizontally in a predetermined order with a separator between each pair of n-best hypotheses to generate the single text data structure, wherein each separator is set based on a classification algorithm of the classifier of the speech recognition system as a symbol that is usable by the classifier of the speech recognition system; and outputting the single text data structure as an input to the classifier to perform a classification task; wherein the classifier is trained to perform the classification task based on a single training text data structure by; obtaining source training data including a plurality of second n-best hypotheses and a transcription for each utterance from a database; arranging the source training data with the transcription at a head or at an end of the second n-best hypotheses depending on the predetermined order to generate the single training text data structure; and outputting the single training text data structure to the classifier. - View Dependent Claims (8, 9)
-
Specification