Recent advances in efficient decoding combining on-line transducer composition and smoothed language model incorporation

Willett,; Katagiri,

doi:10.1109/icassp.2002.1005839

Cited by 5 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They include Dol ng [4], Willett [5], Caseiro [6] and Hori [7]. The rst step of any dynamic composition algorithm is to factorize the entire search space into two or more component WFSTs before decoding.…”

Section: Current Approaches To Dynamic Wfst Compositionmentioning

confidence: 99%

A Generalized Dynamic Composition Algorithm of Weighted Finite State Transducers for Large Vocabulary Speech Recognition

Cheng

Dines

Magimai.-Doss

2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

View full text Add to dashboard Cite

We propose a generalized dynamic composition algorithm of weighted nite state transducers (WFST), which avoids the creation of noncoaccessible paths, performs weight look-ahead and does not impose any constraints to the topology of the WFSTs. Experimental results on Wall Street Journal (WSJ1) 20k-word trigram task show that at 17% WER (moderately-wide beam width), the decoding time of the proposed approach is about 48% and 65% of the other two dynamic composition approaches. In comparison with static composition, at the same level of 17% WER, we observe a reduction of about 60% in memory requirement, with an increase of about 60% in decoding time due to extra overheads for dynamic composition.

show abstract

Section: Current Approaches To Dynamic Wfst Compositionmentioning

confidence: 99%

A Generalized Dynamic Composition Algorithm of Weighted Finite State Transducers for Large Vocabulary Speech Recognition

Cheng

Dines

Magimai.-Doss

2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

View full text Add to dashboard Cite

show abstract

“…One of the possible solutions to this problem is to perform on-the-fly transducer composition during decoding. Acoustical, phonetic and lexical resources may still be composed and optimised off-line, while the language model transducer is locally, dynamically composed at run time [3,19,9]. By using this approach, we can avoid composing part of the search space which is not traversed by any hypotheses.…”

Section: Future Developmentmentioning

confidence: 99%

Juicer: A Weighted Finite-State Transducer Speech Decoder

Moore

Dines

Magimai.-Doss

et al. 2006

Machine Learning for Multimodal Interaction

View full text Add to dashboard Cite

Abstract. A major component in the development of any speech recognition system is the decoder. As task complexities and, consequently, system complexities have continued to increase the decoding problem has become an increasingly significant component in the overall speech recognition system development effort, with efficient decoder design contributing to significantly improve the trade-off between decoding time and search errors. In this paper we present the "Juicer" (from transducer ) large vocabulary continuous speech recognition (LVCSR) decoder based on weighted finite-State transducer (WFST). We begin with a discussion of the need for open source, state-of-the-art decoding software in LVCSR research and how this lead to the development of Juicer, followed by a brief overview of decoding techniques and major issues in decoder design. We present Juicer and its major features, emphasising its potential not only as a critical component in the development of LVCSR systems, but also as an important research tool in itself, being based around the flexible WFST paradigm. We also provide results of benchmarking tests that have been carried out to date, demonstrating that in many respects Juicer, while still in its early development, is already achieving state-of-the-art. These benchmarking tests serve to not only demonstrate the utility of Juicer in its present state, but are also being used to guide future development, hence, we conclude with a brief discussion of some of the extensions that are currently under way or being considered for Juicer.

show abstract

“…In [2], B = G was separated entirely from the other components. In [5] and [19], G was separated into an incremental LM B = Gi as well as a smearing LM Gs which was statically composed with the other components to form A = HC • L • Gs. Finally, in [9], all components were dynamically composed during recognition.…”

Section: Introductionmentioning

confidence: 99%

An algorithm for fast composition of weighted finite-state transducers

McDonough

Stoimenov

Klakow

2007

2007 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding (ASRU)

View full text Add to dashboard Cite

In automatic speech recognition based on weighted-finite transducers, a static decoding graph HC • L • G is typically constructed. In this work, we first show how the size of the decoding graph can be reduced and the necessity of determinizing it can be eliminated by removing the ambiguity associated with transitions to the backoff state or states in G. We then show how the static construction can be avoided entirely by performing fast on-the-fly composition of HC and L • G. We demonstrate that speech recognition based on this on-the-fly composition approximately 80% more run-time than recognition based on the statically-expanded network R, which makes it competitive compared with other dynamic expansion algorithms that have appeared in the literature. Moreover, the dynamic algorithm requires a factor of approximately seven less main memory as the recognition based on the static decoding graph.

show abstract

Recent advances in efficient decoding combining on-line transducer composition and smoothed language model incorporation

Cited by 5 publications

References 0 publications

A Generalized Dynamic Composition Algorithm of Weighted Finite State Transducers for Large Vocabulary Speech Recognition

A Generalized Dynamic Composition Algorithm of Weighted Finite State Transducers for Large Vocabulary Speech Recognition

Juicer: A Weighted Finite-State Transducer Speech Decoder

An algorithm for fast composition of weighted finite-state transducers

Contact Info

Product

Resources

About