no code implementations • ACL (IWSLT) 2021 • Antonios Anastasopoulos, Ondřej Bojar, Jacob Bremerman, Roldano Cattoni, Maha Elbayad, Marcello Federico, Xutai Ma, Satoshi Nakamura, Matteo Negri, Jan Niehues, Juan Pino, Elizabeth Salesky, Sebastian Stüker, Katsuhito Sudoh, Marco Turchi, Alexander Waibel, Changhan Wang, Matthew Wiesner
The evaluation campaign of the International Conference on Spoken Language Translation (IWSLT 2021) featured this year four shared tasks: (i) Simultaneous speech translation, (ii) Offline speech translation, (iii) Multilingual speech translation, (iv) Low-resource speech translation.
no code implementations • ACL (IWSLT) 2021 • Ryo Fukuda, Yui Oka, Yasumasa Kano, Yuki Yano, Yuka Ko, Hirotaka Tokuyama, Kosuke Doi, Sakriani Sakti, Katsuhito Sudoh, Satoshi Nakamura
This paper describes NAIST’s system for the English-to-Japanese Simultaneous Text-to-text Translation Task in IWSLT 2021 Evaluation Campaign.
no code implementations • IWSLT (ACL) 2022 • Ryo Fukuda, Yuka Ko, Yasumasa Kano, Kosuke Doi, Hirotaka Tokuyama, Sakriani Sakti, Katsuhito Sudoh, Satoshi Nakamura
This paper describes NAIST’s simultaneous speech translation systems developed for IWSLT 2022 Evaluation Campaign.
no code implementations • IWSLT (ACL) 2022 • Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura
Simultaneous translation is a task that requires starting translation before the speaker has finished speaking, so we face a trade-off between latency and accuracy.
no code implementations • dialdoc (ACL) 2022 • Yuya Nakano, Seiya Kawano, Koichiro Yoshino, Katsuhito Sudoh, Satoshi Nakamura
Ambiguous questions are generated by eliminating a part of a sentence considering the sentence structure.
no code implementations • WMT (EMNLP) 2021 • Kosuke Takahashi, Yoichi Ishibashi, Katsuhito Sudoh, Satoshi Nakamura
This paper describes our submission to the WMT2021 shared metrics task.
no code implementations • ACL (IWSLT) 2021 • Kosuke Doi, Katsuhito Sudoh, Satoshi Nakamura
This paper describes the construction of a new large-scale English-Japanese Simultaneous Interpretation (SI) corpus and presents the results of its analysis.
no code implementations • IWSLT (EMNLP) 2018 • Kaho Osamura, Takatomo Kano, Sakriani Sakti, Katsuhito Sudoh, Satoshi Nakamura
In this paper, a neural sequence-to-sequence ASR is used as feature processing that is trained to produce word posterior features given spoken utterances.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +4
no code implementations • ACL (IWSLT) 2021 • Ryo Fukuda, Katsuhito Sudoh, Satoshi Nakamura
Recent studies argue that knowledge distillation is promising for speech translation (ST) using end-to-end models.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +4
no code implementations • IWSLT (EMNLP) 2018 • Johanes Effendi, Sakriani Sakti, Katsuhito Sudoh, Satoshi Nakamura
In this paper, we investigate and utilize neural paraphrasing to improve translation quality in neural MT (NMT), which has not yet been much explored.
no code implementations • EACL (HumEval) 2021 • Katsuhito Sudoh, Kosuke Takahashi, Satoshi Nakamura
Our classification-based approach focuses on such errors using several error type labels, for practical machine translation evaluation in an age of neural machine translation.
no code implementations • ICON 2021 • Hour Kaing, Chenchen Ding, Katsuhito Sudoh, Masao Utiyama, Eiichiro Sumita, Satoshi Nakamura
Pretrained multilingual language models have become a key part of cross-lingual transfer for many natural language processing tasks, even those without bilingual information.
no code implementations • ICON 2021 • Kohichi Takai, Gen Hattori, Akio Yoneyama, Keiji Yasuda, Katsuhito Sudoh, Satoshi Nakamura
The proposed method applies the Named Entity (NE) fea-ture vector to Factored Transformer for accurate proper noun translation.
no code implementations • IWSLT (ACL) 2022 • Antonios Anastasopoulos, Loïc Barrault, Luisa Bentivogli, Marcely Zanon Boito, Ondřej Bojar, Roldano Cattoni, Anna Currey, Georgiana Dinu, Kevin Duh, Maha Elbayad, Clara Emmanuel, Yannick Estève, Marcello Federico, Christian Federmann, Souhir Gahbiche, Hongyu Gong, Roman Grundkiewicz, Barry Haddow, Benjamin Hsu, Dávid Javorský, Vĕra Kloudová, Surafel Lakew, Xutai Ma, Prashant Mathur, Paul McNamee, Kenton Murray, Maria Nǎdejde, Satoshi Nakamura, Matteo Negri, Jan Niehues, Xing Niu, John Ortega, Juan Pino, Elizabeth Salesky, Jiatong Shi, Matthias Sperber, Sebastian Stüker, Katsuhito Sudoh, Marco Turchi, Yogesh Virkar, Alexander Waibel, Changhan Wang, Shinji Watanabe
The evaluation campaign of the 19th International Conference on Spoken Language Translation featured eight shared tasks: (i) Simultaneous speech translation, (ii) Offline speech translation, (iii) Speech to speech translation, (iv) Low-resource speech translation, (v) Multilingual speech translation, (vi) Dialect speech translation, (vii) Formality control for speech translation, (viii) Isometric speech translation.
no code implementations • IWSLT 2017 • Mauro Cettolo, Marcello Federico, Luisa Bentivogli, Jan Niehues, Sebastian Stüker, Katsuhito Sudoh, Koichiro Yoshino, Christian Federmann
The IWSLT 2017 evaluation campaign has organised three tasks.
no code implementations • 7 Feb 2024 • Roman Koshkin, Katsuhito Sudoh, Satoshi Nakamura
Decoder-only large language models (LLMs) have recently demonstrated impressive capabilities in text generation and reasoning.
no code implementations • 24 Nov 2023 • Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura
In this work, we propose a novel latency evaluation metric for simultaneous translation called \emph{Average Token Delay} (ATD) that focuses on the duration of partial translations.
no code implementations • 14 Jun 2023 • Yuka Ko, Ryo Fukuda, Yuta Nishikawa, Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura
In this paper, we propose an effective way to train a SimulST model using mixed data of SI and offline.
1 code implementation • 25 Apr 2023 • Ryo Fukuda, Katsuhito Sudoh, Satoshi Nakamura
In this study, we extended SHAS to improve ST translation accuracy and efficiency by splitting speech into shorter segments that correspond to sentences.
no code implementations • 23 Apr 2023 • Jinming Zhao, Yuka Ko, Kosuke Doi, Ryo Fukuda, Katsuhito Sudoh, Satoshi Nakamura
Research has been limited due to the lack of a large-scale training corpus.
no code implementations • 1 Mar 2023 • Yuka Okuda, Katsuhito Sudoh, Seitaro Shinagawa, Satoshi Nakamura
A conversational recommender system (CRS) is a practical application for item recommendation through natural language conversation.
1 code implementation • 11 Feb 2023 • Yoichi Ishibashi, Danushka Bollegala, Katsuhito Sudoh, Satoshi Nakamura
To address this question, we conduct a systematic study of the robustness of discrete prompts by applying carefully designed perturbations into an application using AutoPrompt and then measure their performance in two Natural Language Inference (NLI) datasets.
no code implementations • 22 Nov 2022 • Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura
In this work, we propose a novel latency evaluation metric called Average Token Delay (ATD) that focuses on the end timings of partial translations in simultaneous translation.
1 code implementation • 1 Nov 2022 • Keisuke Toyama, Katsuhito Sudoh, Satoshi Nakamura
Although the well-known MR-to-text E2E dataset has been used by many researchers, its MR-text pairs include many deletion/insertion/substitution errors.
1 code implementation • 24 Oct 2022 • Yoichi Ishibashi, Sho Yokoi, Katsuhito Sudoh, Satoshi Nakamura
In the field of natural language processing (NLP), continuous vector representations are crucial for capturing the semantic meanings of individual words.
1 code implementation • 29 Mar 2022 • Ryo Fukuda, Katsuhito Sudoh, Satoshi Nakamura
We also propose a hybrid method that combines VAD and the above speech segmentation method.
no code implementations • WMT (EMNLP) 2021 • Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura
Simultaneous translation is a task in which translation begins before the speaker has finished speaking, so it is important to decide when to start the translation process.
no code implementations • 29 Jul 2021 • Yui Oka, Katsuhito Sudoh, Satoshi Nakamura
Non-autoregressive neural machine translation (NAT) usually employs sequence-level knowledge distillation using autoregressive neural machine translation (AT) as its teacher model.
1 code implementation • SIGDIAL (ACL) 2021 • Shohei Tanaka, Koichiro Yoshino, Katsuhito Sudoh, Satoshi Nakamura
In order to train the classification model on such training data, we applied the positive/unlabeled (PU) learning method, which assumes that only a part of the data is labeled with positive examples.
no code implementations • COLING 2020 • Koichiro Yoshino, Kana Ikeuchi, Katsuhito Sudoh, Satoshi Nakamura
Spoken language understanding (SLU), which converts user requests in natural language to machine-interpretable expressions, is becoming an essential task.
no code implementations • COLING 2020 • Yui Oka, Katsuki Chousa, Katsuhito Sudoh, Satoshi Nakamura
Since length constraints with exact target sentence lengths degrade translation performance, we add random noise within a certain window size to the length constraints in the PE during the training.
no code implementations • 10 Nov 2020 • Katsuhito Sudoh, Takatomo Kano, Sashi Novitasari, Tomoya Yanagita, Sakriani Sakti, Satoshi Nakamura
This paper presents a newly developed, simultaneous neural speech-to-speech translation system and its evaluation.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +6
no code implementations • 19 Oct 2020 • Dušan Variš, Katsuhito Sudoh, Satoshi Nakamura
We present our work in progress exploring the possibilities of a shared embedding space between textual and visual modality.
2 code implementations • ACL 2020 • Yoichi Ishibashi, Katsuhito Sudoh, Koichiro Yoshino, Satoshi Nakamura
For transferring king into queen in this analogy-based manner, we subtract a difference vector man - woman based on the knowledge that king is male.
no code implementations • ACL 2020 • Kosuke Takahashi, Katsuhito Sudoh, Satoshi Nakamura
Our experiments show that our proposed method using Cross-lingual Language Model (XLM) trained with a translation language modeling (TLM) objective achieves a higher correlation with human judgments than a baseline method that uses only hypothesis and reference sentences.
no code implementations • WS 2020 • Ryo Fukuda, Katsuhito Sudoh, Satoshi Nakamura
This paper describes NAIST{'}s NMT system submitted to the IWSLT 2020 conversational speech translation task.
no code implementations • 27 Nov 2019 • Katsuki Chousa, Katsuhito Sudoh, Satoshi Nakamura
Simultaneous machine translation is a variant of machine translation that starts the translation process before the end of an input.
no code implementations • WS 2019 • Hiroaki Hayashi, Yusuke Oda, Alexandra Birch, Ioannis Konstas, Andrew Finch, Minh-Thang Luong, Graham Neubig, Katsuhito Sudoh
This document describes the findings of the Third Workshop on Neural Generation and Translation, held in concert with the annual conference of the Empirical Methods in Natural Language Processing (EMNLP 2019).
2 code implementations • WS 2019 • Shohei Tanaka, Koichiro Yoshino, Katsuhito Sudoh, Satoshi Nakamura
We propose a novel method for selecting coherent and diverse responses for a given dialogue context.
1 code implementation • 20 Nov 2018 • Ryo Nakamura, Katsuhito Sudoh, Koichiro Yoshino, Satoshi Nakamura
Although generation-based dialogue systems have been widely researched, the response generations by most existing systems have very low diversities.
no code implementations • IWSLT (EMNLP) 2018 • Yuta Nishimura, Katsuhito Sudoh, Graham Neubig, Satoshi Nakamura
By using information from these multiple sources, these systems achieve large gains in accuracy.
no code implementations • 30 Jul 2018 • Katsuki Chousa, Katsuhito Sudoh, Satoshi Nakamura
The proposed loss function encourages an NMT decoder to generate words close to their references in the embedding space; this helps the decoder to choose similar acceptable words when the actual best candidates are not included in the vocabulary due to its size limitation.
no code implementations • WS 2018 • Yuta Nishimura, Katsuhito Sudoh, Graham Neubig, Satoshi Nakamura
This study focuses on the use of incomplete multilingual corpora in multi-encoder NMT and mixture of NMT experts and examines a very simple implementation where missing source translations are replaced by a special symbol <NULL>.
no code implementations • WS 2017 • Yusuke Oda, Katsuhito Sudoh, Satoshi Nakamura, Masao Utiyama, Eiichiro Sumita
This paper describes the details about the NAIST-NICT machine translation system for WAT2017 English-Japanese Scientific Paper Translation Task.
no code implementations • WS 2017 • Makoto Morishita, Yusuke Oda, Graham Neubig, Koichiro Yoshino, Katsuhito Sudoh, Satoshi Nakamura
Training of neural machine translation (NMT) models usually uses mini-batches for efficiency purposes.
no code implementations • 2 Apr 2017 • Junki Matsuo, Mamoru Komachi, Katsuhito Sudoh
One of the most important problems in machine translation (MT) evaluation is to evaluate the similarity between translation hypotheses with different surface forms from the reference, especially at the segment level.
no code implementations • 12 Dec 2016 • Xun Wang, Katsuhito Sudoh, Masaaki Nagata, Tomohide Shibata, Daisuke Kawahara, Sadao Kurohashi
This paper introduces a novel neural network model for question answering, the \emph{entity-based memory network}.
no code implementations • COLING 2016 • Xun Wang, Masaaki Nishino, Tsutomu Hirao, Katsuhito Sudoh, Masaaki Nagata
Existing methods focus on the extraction of key information, but often neglect coherence.
no code implementations • WS 2016 • Katsuhito Sudoh, Masaaki Nagata
This paper presents our Chinese-to-Japanese patent machine translation system for WAT 2016 (Group ID: ntt) that uses syntactic pre-ordering over Chinese dependency structures.
no code implementations • WS 2016 • Shin Kanouchi, Katsuhito Sudoh, Mamoru Komachi
This paper presents an improved lexicalized reordering model for phrase-based statistical machine translation using a deep neural network.