Promising Accurate Prefix Boosting For Sequence-to-sequence ASR

Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F19%3APU134184" target="_blank" >RIV/00216305:26230/19:PU134184 - isvavai.cz</a>
Výsledek na webu
<a href="https://ieeexplore.ieee.org/document/8682782" target="_blank" >https://ieeexplore.ieee.org/document/8682782</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/ICASSP.2019.8682782" target="_blank" >10.1109/ICASSP.2019.8682782</a>

Jazyk výsledku
angličtina
Název v původním jazyce
Promising Accurate Prefix Boosting For Sequence-to-sequence ASR
Popis výsledku v původním jazyce
In this paper, we present promising accurate prefix boosting (PAPB), a discriminative training technique for attention based sequence-tosequence (seq2seq) ASR. PAPB is devised to unify the training and testing scheme effectively. 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 external RNNLM respectively on Wall Street Journal dataset.
Název v anglickém jazyce
Promising Accurate Prefix Boosting For Sequence-to-sequence ASR
Popis výsledku anglicky
In this paper, we present promising accurate prefix boosting (PAPB), a discriminative training technique for attention based sequence-tosequence (seq2seq) ASR. PAPB is devised to unify the training and testing scheme effectively. 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 external RNNLM respectively on Wall Street Journal dataset.

Druh
D - Stať ve sborníku
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt
<a href="/cs/project/LQ1602" target="_blank" >LQ1602: IT4Innovations excellence in science</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění
2019
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Podobné výsledky(10)