Articles

Compared to the voluminous literature of sound change and variation across speakers and generations, much less is understood about change and variation within individuals across the lifespan. In this paper, we present a study of the Cantonese speech of a public figure, Mr. Chun-ying Leung (1954-), recorded over a span …...

BLEU is the most well-known automatic evaluation technology in assessing the performance of machine translation systems. However, BLEU does not know which parts of the NMT translation results are good or bad. This paper describes the automatic evaluation approach of NMT systems by linguistic test points. This approach allows automatic …...

The link between music and language has been a subject of great interest, and evidence suggesting a connection between musical abilities and prosodic processing skills in language is growing. Acoustic fundamental frequency (F0), perceived as pitch, differentiates notes in music and word meaning in lexical tone languages. This study examines …...

In this study, we propose a method for extract- ing free translation examples from bilingual parallel corpora based on an innovative use of attention scores. Preliminary results show that the approach is promising and paraphrases at both sentential and sub-sentential levels cover- ing diverse surface forms could be identified. The …...

While extractive summarization is a well stud- ied problem, it is far from solved. In recent years a large number of interesting and com- plex models have been used to achieve signif- icant improvements in performance. This can easily be attributed to Deep Learning models and dense vector representations but …...

This study investigates the English [ADJ of] construction in academic writing. The target construction represents a subclass of the predicative adjectives (The children are happy.) in English. While the predicative adjectives have long been recognized to function either as a subject complement (e.g., The children are happy.) or object complement …...

Large amount of user-generated data are posted online in social media platforms, in- cluding user preferences, dining and leisure activities, events, news and personal blogs. This resulted in varying efforts to process so- cial media data using NLP and ML algorithms for topic classification, sentiment analysis and detection, and events …...

This paper proposes a new method of instance-based domain adaptation for senti- ment analysis. First, our method defines the likelihood of keywords, through the value of inverse document frequency (IDF), for each word in documents in the target domain. Next, the keyword content rate of a document is calculated using …...

AbstractThis paper describes the USTC-NEL (short for ”National Engineering Laboratory for Speech and Language Information Processing University of science and technology of china”) system to the speech translation task of the IWSLT Evaluation 2018. The system is a conventional pipeline system which contains 3 modules: speech recognition, post-processing and machine …...

AbstractIn this paper we present the ADAPT system built for the Basque to English Low Resource MT Evaluation Campaign. Basque is a low-resourced, morphologically-rich language. This poses a challenge for Neural Machine Translation models which usually achieve better performance when trained with large sets of data. Accordingly, we used synthetic …...

AbstractParaphrasing has been proven to improve translation quality in machine translation (MT) and has been widely studied alongside with the development of statistical MT (SMT). In this paper, we investigate and utilize neural paraphrasing to improve translation quality in neural MT (NMT), which has not yet been much explored. Our …...

AbstractData selection techniques applied to neural machine translation (NMT) aim to increase the performance of a model by retrieving a subset of sentences for use as training data. One of the possible data selection techniques are transductive learning methods, which select the data based on the test set, i.e. the …...

AbstractA spoken language translation (ST) system consists of at least two modules: an automatic speech recognition (ASR) system and a machine translation (MT) system. In most cases, an MT is only trained and optimized using error-free text data. If the ASR makes errors, the translation accuracy will be greatly reduced. …...

AbstractMost modern neural machine translation (NMT) systems rely on presegmented inputs. Segmentation granularity importantly determines the input and output sequence lengths, hence the modeling depth, and source and target vocabularies, which in turn determine model size, computational costs of softmax normalization, and handling of out-of-vocabulary words. However, the current practice …...

AbstractThis paper describes the Johns Hopkins University (JHU) and Kyoto University submissions to the Speech Translation evaluation campaign at IWSLT2018. Our end-to-end speech translation systems are based on ESPnet and implements an attention-based encoder-decoder model. As comparison, we also experiment with a pipeline system that uses independent neural network systems …...

AbstractMultilingual neural machine translation (M-NMT) has recently shown to improve performance of machine translation of low-resource languages. Thanks to its implicit transfer-learning mechanism, the availability of a highly resourced language pair can be leveraged to learn useful representation for a lower resourced language. This work investigates how a low-resource translation …...

AbstractThis paper describes FBK’s submission to the end-to-end English-German speech translation task at IWSLT 2018. Our system relies on a state-of-the-art model based on LSTMs and CNNs, where the CNNs are used to reduce the temporal dimension of the audio input, which is in general much higher than machine translation …...

AbstractWe present our submission to the IWSLT18 Low Resource task focused on the translation from Basque-to-English. Our submission is based on the current state-of-the-art self-attentive neural network architecture, Transformer. We further improve this strong baseline by exploiting available monolingual data using the back-translation technique. We also present further improvements gained …...

AbstractThis work describes the En→De Alibaba speech translation system developed for the evaluation campaign of the International Workshop on Spoken Language Translation (IWSLT) 2018. In order to improve ASR performance, multiple ASR models including conventional and end-to-end models are built, then we apply model fusion in the final step. ASR …...

AbstractThis paper describes KIT’s submission to the IWSLT 2018 Translation task. We describe a system participating in the baseline condition and a system participating in the end-to-end condition. The baseline system is a cascade of an ASR system, a system to segment the ASR output and a neural machine translation …...