Promising Accurate Prefix Boosting for sequence-to-sequence ASR

arxiv: 1811.02770 · v1 · pith:FQ7LCYJUnew · submitted 2018-11-07 · 📡 eess.AS · cs.CL· cs.LG· cs.SD

Promising Accurate Prefix Boosting for sequence-to-sequence ASR

Murali Karthick Baskar , Luk\'a\v{s} Burget , Shinji Watanabe , Martin Karafi\'at , Takaaki Hori , Jan Honza \v{C}ernock\'y This is my paper

classification 📡 eess.AS cs.CLcs.LGcs.SD

keywords trainingpapbaccurateboostingprefixpromisingsequence-to-sequenceachieving

0 comments p. Extension

pith:FQ7LCYJU Add to your LaTeX paper

What is a Pith Number?

\usepackage{pith}
\pithnumber{FQ7LCYJU}

Prints a linked pith:FQ7LCYJU badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

read the original abstract

In this paper, we present promising accurate prefix boosting (PAPB), a discriminative training technique for attention based sequence-to-sequence (seq2seq) ASR. PAPB is devised to unify the training and testing scheme in an effective manner. The training procedure involves maximizing the score of each partial correct sequence obtained during beam search compared to other hypotheses. The training objective also includes minimization of token (character) error rate. PAPB shows its efficacy by achieving 10.8\% and 3.8\% WER with and without RNNLM respectively on Wall Street Journal dataset.

This paper has not been read by Pith yet.

Promising Accurate Prefix Boosting for sequence-to-sequence ASR

discussion (0)