Spelling-Aware Word-Based End-to-End ASR

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F22%3APU145323" target="_blank" >RIV/00216305:26230/22:PU145323 - isvavai.cz</a>
Result on the web
<a href="https://ieeexplore.ieee.org/document/9833231" target="_blank" >https://ieeexplore.ieee.org/document/9833231</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/LSP.2022.3192199" target="_blank" >10.1109/LSP.2022.3192199</a>

Result language
angličtina
Original language name
Spelling-Aware Word-Based End-to-End ASR
Original language description
We propose a new end-to-end architecture for automatic speech recognition that expands the listen, attend and spell (LAS) paradigm. While the main word-predicting network is trained to predict words, the secondary, speller network, is optimized to predict word spellings from inner representations of the main network (e.g. word embeddings or context vectors from the attention module). We show that this joint training improves the word error rate of a word-based system and enables solving additional tasks, such as out-of-vocabulary word detection and recovery. The tests are conducted on LibriSpeech dataset consisting of 1000h of read speech.
Czech name
—
Czech description
—

Type
J<sub>imp</sub> - Article in a specialist periodical, which is included in the Web of Science database
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)