Spoken-text normalization that converts spoken-style text into style normalized text is becoming an important technology for improving subsequent processing such as machine translation andsummarization . MAsked Pointer-Generator Network(MAPGN) is a novel self-supervised learning method . MAPGN is more effective for pointer-generatornetworks than the conventional methods in twospoken text normalization tasks . The proposed method can effectively pre-train the pointers-generator network by learning to fill masked tokens using the copy mechanism . Ourexperiments demonstrate that MAPGN was more effective than conventional methods to learn to fill masks using a copy mechanism in a speech-style normalization task than it was used to train a pointer-generation network that can be used to fill masked tokens . The MAPGN method is

Author(s) : Mana Ihori, Naoki Makishima, Tomohiro Tanaka, Akihiko Takashima, Shota Orihashi, Ryo Masumura

Links : PDF - Abstract

Code :

https://github.com/oktantod/RoboND-DeepLearning-Project


Coursera

Keywords : mapgn - pointer - text - network - masked -

Leave a Reply

Your email address will not be published. Required fields are marked *