Paper Info
Title
Acoustically discriminative language model training with pseudo-hypothesis
Abstract
Recently proposed methods for discriminative language modeling require alternate hypotheses in the form of lattices or N-best lists. These are usually generated by an Automatic Speech Recognition (ASR) system on the same speech data used to train the system. This requirement restricts the scope of these methods to corpora where both the acoustic material and the corresponding true transcripts are available. Typically, the text data available for language model (LM) training is an order of magnitude larger than manually transcribed speech. This paper provides a general framework to take advantage of this volume of textual data in the discriminative training of language models. We propose to generate probable N-best lists directly from the text material, which resemble the N-best lists produced by an ASR system by incorporating phonetic confusability estimated from the acoustic model of the ASR system. We present experiments with Japanese spontaneous lecture speech data, which demonstrate that discriminative LM training with the proposed framework is effective and provides modest gains in ASR accuracy.
Year
DOI
Venue
2012
10.1016/j.specom.2011.08.004
Speech Communication
Keywords
Field
DocType
asr system,acoustically discriminative language model,textual data,language model,discriminative language modeling,n-best list,speech data,text data,asr accuracy,japanese spontaneous lecture speech,discriminative lm training,finite state transducer
Computer science,Speech recognition,Natural language processing,Artificial intelligence,Discriminative model,Finite state transducer,Language model,Acoustic model
Journal
Volume
Issue
ISSN
54
2
0167-6393
Citations 
PageRank 
References 
3
0.39
26
Authors
6
Name
Order
Citations
PageRank
Gakuto Kurata110719.06
Abhinav Sethy236331.16
Bhuvana Ramabhadran31779153.83
Ariya Rastrow424323.49
Nobuyasu Itoh56513.19
Masafumi Nishimura611222.77