Using Neural Machine Translation for Normalising Historical Documents thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Using Neural Machine Translation for Normalising Historical Documents

Published on Nov 05, 20244 Views

Session 2: Lexicography and Language Technologies

Related categories

Chapter list

USING NEURAL MACHINE TRANSLATION FOR NORMALISING HISTORICAL DOCUMENTS00:00
INTRODUCTION00:20
HISTORICAL DOCUMENT SPELLING01:09
INTRODUCTION02:12
INTRODUCTION03:25
HISTORICAL CORPUS: PORTUGUESE04:30
HISTORICAL CORPUS: ALBANIAN04:52
TRAIN, DEVELOPMENT AND TEST SETS: PORTUGUESE05:39
TRAIN, DEVELOPMENT AND TEST SETS: ALBANIAN06:10
NMT MODELS06:19
BLUE-SCORE RESULTS ON PORTUGUESE AND ALBANIAN07:21
BLUE-SCORE RESULTS ON PORTUGUESE AND ALBANIAN08:12
BLUE-SCORE RESULTS ON PORTUGUESE AND ALBANIAN08:17
DISCUSSION08:57
LIMITATIONS10:19
CONTRIBUTIONS AND FINAL REMARKS12:36
THANK YOU VERY MUCH !13:20