Attention is the core mechanism of today's most used architectures for
n...
The study of the attention mechanism has sparked interest in many fields...
Modern automatic translation systems aim at place the human at the cente...
Automatic subtitling is the task of automatically translating the speech...
Speech translation for subtitling (SubST) is the task of automatically
t...
Simultaneous speech translation (SimulST) systems aim at generating thei...
Recent work has shown that systems for speech translation (ST) – similar...
The primary goal of this FBK's systems submission to the IWSLT 2022 offl...
In simultaneous speech translation (SimulST), finding the best trade-off...
Gender bias is largely recognized as a problematic phenomenon affecting
...
Simultaneous speech translation (SimulST) is the task in which output
ge...
Automatic translation systems are known to struggle with rare words. Amo...
Transformer-based models have gained increasing popularity achieving
sta...
With the increased audiovisualisation of communication, the need for liv...
Speech translation (ST) has lately received growing interest for the
gen...
This paper describes FBK's system submission to the IWSLT 2021 Offline S...
Five years after the first published proofs of concept, direct approache...
Having recognized gender bias as a major issue affecting current transla...
The audio segmentation mismatch between training data and those seen at
...
Machine translation (MT) technology has facilitated our daily tasks by
p...
Neural Machine Translation (NMT) approaches employing monolingual data a...
We present the Multilingual TEDx corpus, built to support speech recogni...
Previous studies demonstrated that a dynamic phone-informed compression ...
Direct speech translation (ST) has shown to be a complex task requiring
...
In automatic speech translation (ST), traditional cascade approaches
inv...
Recent studies on direct speech translation show continuous improvements...
Direct speech-to-text translation (ST) models are usually trained on cor...
Translating from languages without productive grammatical gender like En...
This paper describes FBK's participation in the IWSLT 2020 offline speec...
Subtitling is becoming increasingly important for disseminating informat...
Recent advents in Neural Machine Translation (NMT) have shown improvemen...
Growing needs in localising audiovisual content in multiple languages th...
Multilingual Neural Machine Translation (MNMT) for low-resource language...
Despite recent technology advancements, the effectiveness of neural
appr...
Nowadays, training end-to-end neural models for spoken language translat...
We propose a neural machine translation (NMT) approach that, instead of
...
In recent years, Neural Machine Translation (NMT) has been shown to be m...
Recent work on multilingual neural machine translation reported competit...
We propose a method to transfer knowledge across neural machine translat...
This paper describes FBK's submission to the end-to-end English-German s...
Training models for the automatic correction of machine-translated text
...
The necessity of using a fixed-size word vocabulary in order to control ...
Recognizer Output Voting Error Reduction (ROVER) has been widely used fo...
In this paper we propose to exploit the automatic Quality Estimation (QE...
Deriving prior polarity lexica for sentiment analysis - where positive o...