Filters








101 Hits in 5.4 sec

Improving the Robustness of Speech Translation [article]

Xiang Li, Haiyang Xue, Wei Chen, Yang Liu, Yang Feng, Qun Liu
2018 arXiv   pre-print
Besides, we also incorporate the Chinese Pinyin feature which is easy to get in speech translation to further improve the translation performance.  ...  on the WMT'17 Chinese-English test set.  ...  single neural network (Chiu et al. 2017) .  ... 
arXiv:1811.00728v1 fatcat:2bpgc2nrnjcajc2jwwq2k6kbxe

Cross-lingual Transfer Learning for

Andrew Johnson, Penny Karanasou, Judith Gaspers, Dietrich Klakow
2019 Proceedings of the 2019 Conference of the North  
A deep neural network model is adopted and the best combination of weights to transfer is extensively investigated.  ...  This work explores cross-lingual transfer learning (TL) for named entity recognition, focusing on bootstrapping Japanese from English.  ...  Working with neural network-based models, this is achieved by initializing some layers of the target network using the weights of the source network, which is assumed to be already trained using a (large  ... 
doi:10.18653/v1/n19-2023 dblp:conf/naacl/JohnsonKGK19 fatcat:fi4mzz7w65gnpoiqvpo2z4j76i

Sub-Character Tokenization for Chinese Pretrained Language Models [article]

Chenglei Si, Zhengyan Zhang, Yingfa Chen, Fanchao Qi, Xiaozhi Wang, Zhiyuan Liu, Yasheng Wang, Qun Liu, Maosong Sun
2021 arXiv   pre-print
Tokenization is fundamental to pretrained language models (PLMs). Existing tokenization methods for Chinese PLMs typically treat each character as an indivisible token.  ...  homophones into the same transliteration sequences and produce the same tokenization output, hence being robust to all homophone typos.  ...  , neural, or contradiction. so that it provides an apple-to-apple comparison CHID (Zheng et al., 2019) is a cloze-style multiple- with our proposed methods.  ... 
arXiv:2106.00400v2 fatcat:rf4u6indwnas3elrc77sqb3hwe

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

Jose Camacho-Collados, Mohammad Taher Pilehvar
2018 The Journal of Artificial Intelligence Research  
, adaptability to different domains and compositionality.  ...  Over the past years, distributed semantic representations have proved to be effective and flexible keepers of prior knowledge to be integrated into downstream applications.  ...  Acknowledgments The authors wish to thank the anonymous reviewers for their comments which helped improve the overall quality of this survey.  ... 
doi:10.1613/jair.1.11259 fatcat:e6bjteqkdjdjjf3x7g56vcbjha

From Word to Sense Embeddings: A Survey on Vector Representations of Meaning [article]

Jose Camacho-Collados, Mohammad Taher Pilehvar
2018 arXiv   pre-print
, adaptability to different domains and compositionality.  ...  Over the past years, distributed semantic representations have proved to be effective and flexible keepers of prior knowledge to be integrated into downstream applications.  ...  Acknowledgments The authors wish to thank the anonymous reviewers for their comments which helped improve the overall quality of this survey.  ... 
arXiv:1805.04032v3 fatcat:eg2omwi2zbb4hpz4eyp4su5ire

The Taxonomy of Writing Systems: How to Measure how Logographic a System is

Richard Sproat, Alexander Gutkin
2021 Computational Linguistics  
We compare this with a simple lexical measure, and an entropic measure, as well as several other neural models, and argue that on balance our attention-based measure accords best with intuition about how  ...  In an ideal phonographic system, the model should need to attend to only the current token in order to compute how to spell it, and this would show in the attention matrix activations.  ...  Acknowledgments We thank Kyle Gorman, Brian Roark and Terry Joyce for helpful comments on an earlier version of this paper.  ... 
doi:10.1162/coli_a_00409 fatcat:jftmdlwajncpvk2law5pk62oey

Pinyin as Subword Unit for Chinese-Sourced Neural Machine Translation

Jinhua Du, Andy Way
2017 Irish Conference on Artificial Intelligence and Cognitive Science  
In this paper, we propose to utilize Pinyin, a romanization system for Chinese characters, to convert Chinese characters to subword units to alleviate the UNK problem.  ...  For alphabetic languages such as English, German and French, transforming a word into subwords is an effective way to alleviate the UNK problem, such as the Byte Pair encoding (BPE) algorithm.  ...  We would like to thank the reviewers for their valuable and constructive comments.  ... 
dblp:conf/aics/DuW17 fatcat:qyurg3y6o5fuxf67p3nynouwia

Survey of Automatic Spelling Correction

Daniel Hládek, Ján Staš, Matúš Pleva
2020 Electronics  
The second group uses an additional model of context. The third group of automatic spelling correction systems in the survey can adapt its model to the given problem.  ...  Although each article contains a brief introduction to the topic, there is a lack of work that would summarize the theoretical framework and provide an overview of the approaches developed so far.  ...  In our opinion, the reason is that most of the spelling approaches strongly depend on the specifics of the language and are hard to adapt to another language or a different application.  ... 
doi:10.3390/electronics9101670 fatcat:pgf65dpwp5b2xc2hc6xxf5pplm

Generating the Voice of the Interactive Virtual Assistant [chapter]

Adriana Stan, Beáta Lőrincz
2021 Virtual Assistant [Working Title]  
This chapter introduces an overview of the current approaches for generating spoken content using text-to-speech synthesis (TTS) systems, and thus the voice of an Interactive Virtual Assistant (IVA).  ...  The speech synthesis methodologies' description begins with the basic, easy to run, low-requirement rule-based synthesis, and ends up within the state-of-the-art deep learning landscape.  ...  The POS is important to disambiguate non-homophone homographs.  ... 
doi:10.5772/intechopen.95510 fatcat:h6dyuvgkk5awtjt4iqbmdhznhe

Compositional semantics network with multi-task learning for pun location

Junyu Mao, Rongbo Wang, Xiaoxi Huang, Zhiqun Chen
2020 IEEE Access  
We present an approach that considers long-distance and short-distance semantic relations between words simultaneously.  ...  We introduce it as an auxiliary to jointly train the original pun location task, which first learns the location of different types of puns together.  ...  The Dalian University of Technology proposed a Chinese humour type recognition task to distinguish homophone, heterography and reversal in CCL2018. Multi-task learning is also related to our work.  ... 
doi:10.1109/access.2020.2978208 fatcat:3g62cx32dvdvddum42rqzusgsi

Message from the general chair

Benjamin C. Lee
2015 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)  
To maximize the utility of the injected knowledge, we deploy a learning-based multi-sieve approach and develop novel entity-based features.  ...  methods as an isolated inference procedure at the end.  ...  Fast Online Training with Frequency-Adaptive Learn- ing Rates for Chinese Word Segmentation and New Word Detection X.  ... 
doi:10.1109/ispass.2015.7095776 dblp:conf/ispass/Lee15 fatcat:ehbed6nl6barfgs6pzwcvwxria

Multimodal Machine Translation through Visuals and Speech

Umut Sulubacak, Ozan Caglayan, Stig-Arne Grönroos, Aku Rouhe, Desmond Elliott, Lucia Specia, Jörg Tiedemann
2019 Zenodo  
This survey reviews the major data resources for these tasks, the evaluation campaigns concentrated around them, the state of the art in end-to-end and pipeline approaches, and also the challenges in performance  ...  The paper concludes with a discussion of directions for future research in these areas: the need for more expansive and challenging datasets, for targeted evaluations of model performance, and for multimodality  ...  We would also like to thank Maarit Koponen for her valuable feedback and her help in establishing our discussions of machine translation evaluation.  ... 
doi:10.5281/zenodo.3690791 fatcat:otdy5i33fzfsnnbb3xgb6zph6q

Multimodal machine translation through visuals and speech

Umut Sulubacak, Ozan Caglayan, Stig-Arne Grönroos, Aku Rouhe, Desmond Elliott, Lucia Specia, Jörg Tiedemann
2020 Machine Translation  
This survey reviews the major data resources for these tasks, the evaluation campaigns concentrated around them, the state of the art in end-to-end and pipeline approaches, and also the challenges in performance  ...  The paper concludes with a discussion of directions for future research in these areas: the need for more expansive and challenging datasets, for targeted evaluations of model performance, and for multimodality  ...  We would also like to thank Maarit Koponen for her valuable feedback and her help in establishing our discussions of machine translation evaluation.  ... 
doi:10.1007/s10590-020-09250-0 fatcat:jod3ghcsnnbipotcqp6sme4lna

What is Needed for a Robot to Acquire Grammar? Some Underlying Primitive Mechanisms for the Synthesis of Linguistic Ability

C. Lyon, Y. Sato, J. Saunders, C.L. Nehaniv
2009 IEEE Transactions on Autonomous Mental Development  
It focuses on issues arising from the use of real language with all its evolutionary baggage, in contrast to an artificial communication system, and describes approaches to addressing these issues.  ...  An overview is given of our own initial experiments in which a robot acquires some basic linguistic capacity via interacting with a human.  ...  Consider "I want two sweets", "I want to go", "I want to too". Serial processing enables homophones to be disambiguated.  ... 
doi:10.1109/tamd.2009.2037731 fatcat:gkpdghr6lbfunnjopnnxpvfjyu

Deep Speech 2: End-to-End Speech Recognition in English and Mandarin [article]

Dario Amodei, Rishita Anubhai, Eric Battenberg, Carl Case, Jared Casper, Bryan Catanzaro, Jingdong Chen, Mike Chrzanowski, Adam Coates, Greg Diamos, Erich Elsen, Jesse Engel (+21 others)
2015 arXiv   pre-print
We show that an end-to-end deep learning approach can be used to recognize either English or Mandarin Chinese speech--two vastly different languages.  ...  Because it replaces entire pipelines of hand-engineered components with neural networks, end-to-end learning allows us to handle a diverse variety of speech including noisy environments, accents and different  ...  Acknowledgments We are grateful to Baidu's speech technology group for help with data preparation and useful conversations.  ... 
arXiv:1512.02595v1 fatcat:auol4dnoxrc5rmj2yrf2kxt5ya
« Previous Showing results 1 — 15 out of 101 results