Machine Translation Approach for Vietnamese Diacritic Restoration

Diep, Thi Ngoc; Nguyen, Duy Binh; Mae, Dang Khoa; Tran, Do Dat

doi:10.1109/ialp.2013.30

Cited by 4 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine translation-based approach has emerged as the best way to handle this problem. [1] and [7] formulate this task as machine translation problem and apply phrase-based translation method by using Moses toolkit. Both of them report the accuracy of 99% on their dataset but the size of this dataset is relativity small.…”

Section: Related Workmentioning

confidence: 99%

“…The phrase translation model is based on the noisy channel model. In particular, it tries to maximize the translation probability from source sentence f to target sentence e. Applying Bayes rule, we can reformulate this probability as arg max e p(e|f) = arg max e p(f|e)p(e)ω length(e) (1) where ω is added to calibrate the output length.…”

Section: A Phrase-based Machine Translationmentioning

confidence: 99%

“…In particular, people need to install some Vietnamese keyboard applications and follow some typing rules such as Telex, VNI, or VIQR to type Vietnamese texts with diacritic marks. According to the statistics in [1], 95% Vietnamese words contain diacritic marks and 80% of these words are ambiguous when removing diacritic marks from them. Thus, reading non-diacritic is difficult for both human and machine.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the use of machine translation-based approaches for vietnamese diacritic restoration

Pham¹,

Pham

Le-Hong

2017

2017 International Conference on Asian Language Processing (IALP)

View full text Add to dashboard Cite

This paper presents an empirical study of two machine translation-based approaches for Vietnamese diacritic restoration problem, including phrase-based and neural-based machine translation models. This is the first work that applies neural-based machine translation method to this problem and gives a thorough comparison to the phrase-based machine translation method which is the current state-of-the-art method for this problem. On a large dataset, the phrase-based approach has an accuracy of 97.32% while that of the neural-based approach is 96.15%. While the neural-based method has a slightly lower accuracy, it is about twice faster than the phrase-based method in terms of inference speed. Moreover, neural-based machine translation method has much room for future improvement such as incorporating pre-trained word embeddings and collecting more training data.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Phrase-based Machine Translationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation