Contextualizing ASR Lattice Rescoring with Hybrid Pointer Network Language Model

Liu Da-Rong
Liu Da-Rong
Liu Chunxi
Liu Chunxi
Zhang Frank
Zhang Frank
Synnaeve Gabriel
Synnaeve Gabriel
Saraf Yatharth
Saraf Yatharth

INTERSPEECH, pp. 3650-3654, 2020.

Cited by: 0|Bibtex|Views45|DOI:https://doi.org/10.21437/Interspeech.2020-1344
EI
Other Links: arxiv.org|dblp.uni-trier.de|academic.microsoft.com

Abstract:

Videos uploaded on social media are often accompanied with textual descriptions. In building automatic speech recognition (ASR) systems for videos, we can exploit the contextual information provided by such video metadata. In this paper, we explore ASR lattice rescoring by selectively attending to the video descriptions. We first use an...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments