Preliminary Evaluations of a WFST Speech Decoder
نویسندگان
چکیده
In this paper we present preliminary evaluations on the large vocabulary speech decoder we are currently developing at Tokyo Institute of Technology. Our goal is to build a scalable and flexible decoder to operate on weighted finite state transducer (WFST) search spaces. Even though the development of the decoder is still in its infancy we are already achieving good accuracy and speed on a large vocabulary spontaneous speech task.
منابع مشابه
An Empirical Comparison of the T , Juicer, HDecode and Sphinx3 Decoders
In this paper we perform a cross-comparison of the T 3 WFST decoder against three different speech recognition decoders on three separate tasks of variable difficulty. We show that the T 3 decoder performs favorably against several established veterans in the field, including the Juicer WFST decoder, Sphinx3, and HDecode in terms of RTF versus Word Accuracy. In addition to comparing decoder per...
متن کاملAn empirical comparison of the t3, juicer, HDecode and sphinx3 decoders
In this paper we perform a cross-comparison of the T 3 WFST decoder against three different speech recognition decoders on three separate tasks of variable difficulty. We show that the T 3 decoder performs favorably against several established veterans in the field, including the Juicer WFST decoder, Sphinx3, and HDecode in terms of RTF versus Word Accuracy. In addition to comparing decoder per...
متن کاملInvestigation on the effects of ASR tuning on speech translation performance
In this paper we describe some of our recent investigations into ASR and SMT coupling issues from an ASR perspective. Our study was motivated by several areas: Firstly, to understand how standard ASR tuning procedures effect the SMT performance and whether it is safe to perform this tuning in isolation. Secondly, to investigate how vocabulary and segmentation mismatches between the ASR and SMT ...
متن کاملOpen-Source WFST tools for LVCSR Cascade Construction
This paper introduces an open source toolkit, Transducersaurus [1], leveraging OpenFst [2] which can be used to build WFST-based networks for LVCSR tasks. The toolkit provides a set of classes for generating each of the fundamental components of a typical WFST ASR cascade, including HMM, context-dependency, lexicon, language model and silence class models. The toolkit further implements a simpl...
متن کاملJuicer: A Weighted Finite-State Transducer Speech Decoder
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 dec...
متن کامل