Proceedings Article10.1109/ASRU.2009.5373438
Iterative decoding: A novel re-scoring framework for confusion networks
Anoop Deoras,Frederick Jelinek +1 more
- 01 Dec 2009
- pp 282-286
TL;DR: Experiments with Language Model re-scoring show that for comparable performance (in terms of word error rate (WER)) of Iterative Decoding and N-best list re- scoring, the search effort required by the method is 22 times less than that of the N- best list method.
read more
Abstract: Recently there has been a lot of interest in confusion network re-scoring using sophisticated and complex knowledge sources. Traditionally, re-scoring has been carried out by the N-best list method or by the lattices or confusion network dynamic programming method. Although the dynamic programming method is optimal, it allows for the incorporation of only Markov knowledge sources. N-best lists, on the other hand, can incorporate sentence level knowledge sources, but with increasing N, the re-scoring becomes computationally very intensive. In this paper, we present an elegant framework for confusion network re-scoring called ‘Iterative Decoding’. In it, integration of multiple and complex knowledge sources is not only easier but it also allows for much faster re-scoring as compared to the N-best list method. Experiments with Language Model re-scoring show that for comparable performance (in terms of word error rate (WER)) of Iterative Decoding and N-best list re-scoring, the search effort required by our method is 22 times less than that of the N-best list method.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Approximate Inference: A Sampling Based Modeling Technique to Capture Complex Dependencies in a Language Model
Anoop Deoras,Tomas Mikolov,Stefan Kombrink,Kenneth Church +3 more
- 01 Aug 2012
TL;DR: The authors used variational approximations of the long-span and complex language models for the first pass decoding and also for the second pass lattice re-scoring in speech recognition systems.
Hill climbing on speech lattices: A new rescoring framework
Ariya Rastrow,Markus Dreyer,Abhinav Sethy,Sanjeev Khudanpur,Bhuvana Ramabhadran,Mark Dredze +5 more
- 22 May 2011
TL;DR: This work describes a new approach for rescoring speech lattices that does not entail computationally intensive lattice expansion or limited rescoring of only an N-best list, and demonstrates empirically that to achieve the same reduction in error rate using a better estimated, higher order language model, this technique evaluates fewer utterance-length hypotheses than conventional N- best rescoring by two orders of magnitude.
Model combination for Speech Recognition using Empirical Bayes Risk minimization
Anoop Deoras,Denis Filimonov,Mary P. Harper,Frederick Jelinek +3 more
- 01 Dec 2010
TL;DR: This paper uses minimum Empirical Bayes Risk for the optimization criterion and Deterministic Annealing techniques to search through the non-convex parameter space to solve the model combination problem for rescoring Automatic Speech Recognition hypotheses.
Improving the Readability of ASR Results for Lectures Using Multiple Hypotheses and Sentence-Level Knowledge
TL;DR: A novel algorithm is proposed that infers clean, readable transcripts from spontaneous multiple hypotheses represented by a confusion network while integrating sentence-level knowledge.
5
Efficient discriminative training of long-span language models
Ariya Rastrow,Mark Dredze,Sanjeev Khudanpur +2 more
- 01 Dec 2011
TL;DR: This work presents discrim inative hill climbing, an efficient and effective discriminative training procedure for long-span LMs based on a hill climbing rescoring algorithm and empirically demonstrates significant computational savings as well as error-rate reduction over N-best training methods in a state of the art ASR system for Broadcast News transcription.
References
A Maximum Likelihood Approach to Continuous Speech Recognition
TL;DR: This paper describes a number of statistical models for use in speech recognition, with special attention to determining the parameters for such models from sparse data, and describes two decoding methods appropriate for constrained artificial languages and one appropriate for more realistic decoding tasks.
1.7K
•Proceedings Article
Explicit word error minimization in N-Best list rescoring
Andreas Stolcke,Yochai Konig,Mitchel Weintraub +2 more
- 01 Jan 1997
TL;DR: A new algorithm is developed that explicitly minimizes expected word error for recognition hypotheses, and approximate the posterior hypothesis probabilities using N-best lists and chooses the hypothesis with the lowest error.
196
Finding consensus in speech recognition
Eric D. Brill,Lidia Mangu +1 more
- 01 Jan 2000
TL;DR: This thesis explores new ways of utilizing the information existing in word lattices produced by speech recognition systems to improve the accuracy of the recognition output and obtain a more perspicuous representation of a set of alternative hypotheses.
62
Error corrective mechanisms for speech recognition
Lidia Mangu,Mukund Padmanabhan +1 more
- 07 May 2001
TL;DR: The paper uses transformation-based learning for inducing a set of rules to guide a better decision between the top two candidates with the highest posterior probabilities in each confusion set, and shows significant improvements over the consensus decoding approach.
48
•Proceedings Article
SRILM – An Extensible Language Modeling Toolkit
Andreas Stolcke
- 01 Jan 2002
TL;DR: The functionality of the SRILM toolkit is summarized and its design and implementation is discussed, highlighting ease of rapid prototyping, reusability, and combinability of tools.
Related Papers (5)
Ciprian Chelba,Frederick Jelinek +1 more
Denis Filimonov,Mary P. Harper +1 more
- 06 Aug 2009
Peter Beyerlein
- 12 May 1998