Search Results for author:
Found 19 papers, 5 papers with code
Tilde's Parallel Corpus Filtering Methods for WMT 2018
The paper describes parallel corpus filtering methods that allow reducing noise of noisy {``}parallel{''} corpora from a level where the corpora are not usable for neural machine translation training (i. e., the resulting systems fail to achieve reasonable translation quality; well below 10 BLEU points) up to a level where the trained systems show decent (over 20 BLEU points on a 10 million word dataset and up to 30 BLEU points on a 100 million word dataset).
Tilde's Machine Translation Systems for WMT 2018
For the WMT 2018 shared task, we submitted seven systems (both constrained and unconstrained) for English-Estonian and Estonian-English translation directions.
Ranked #1 on
Machine Translation
on WMT 2018 Estonian-English
NMT or SMT: Case Study of a Narrow-domain English-Latvian Post-editing Project
The recent technological shift in machine translation from statistical machine translation (SMT) to neural machine translation (NMT) raises the question of the strengths and weaknesses of NMT.
Designing a Speech Corpus for the Development and Evaluation of Dictation Systems in Latvian
In this paper the authors present a speech corpus designed and created for the development and evaluation of dictation systems in Latvian.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Designing the Latvian Speech Recognition Corpus
In this paper the authors present the first Latvian speech corpus designed specifically for speech recognition purposes.
Terminology localization guidelines for the national scenario
Analysis of the terms proves that, in general, in the normative terminology work in Latvia localized terms are coined according to these guidelines.
Bilingual dictionaries for all EU languages
In this work we present three different methods for cleaning noise from automatically generated bilingual dictionaries: LLR, pivot and translation based approach.
Collecting and Using Comparable Corpora for Statistical Machine Translation
Lack of sufficient parallel data for many languages and domains is currently one of the major obstacles to further advancement of automated translation.
Latvian and Lithuanian Named Entity Recognition with TildeNER
It also gives evaluation on human annotated gold standard test corpora for Latvian and Lithuanian languages as well as comparative performance analysis to a state-of-the art English named entity recognition system using parallel and strongly comparable corpora.
