Transfer-based statistical translation of Taiwanese sign language using PCFG

Wu, Chung-Hsien; Su, Hung-Yu; Chiu, Yu-Wen; Lin, Chia‐Hung

doi:10.1145/1227850.1227851

Cited by 17 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here again, there is limited work to be found on SL data-driven approaches, compared to the large literature on text, though the newer attempts fall mostly in this category, see for example [22,13,16,5,18]. A hybrid approach was proposed, falling back on example search when pure statistical learning fails [21].…”

Section: Traditional and Present Methodsmentioning

confidence: 99%

A rule triggering system for automatic text-to-sign translation

Filhol

Hadjadj

Testu

2015

Univ Access Inf Soc

View full text Add to dashboard Cite

The topic of this paper is machine translation (MT) from French text to French Sign Language (LSF). After arguing in favour of a rule-based method, it presents the architecture of an original MT system, built on two distinct efforts: formalising LSF production rules and triggering them with text processing. The former is made without any concern for text or translation and involves corpus analysis to link LSF form features to linguistic functions. It produces a set of production rules which may constitute a full LSF production grammar. The latter is an information extraction task from text, broken down in as many subtasks as there are rules in the grammar. After discussing this architecture, comparing it to the traditional methods and presenting the methodology for each task, the paper present the set of production rules found to govern event precedence and duration in LSF, and gives a progress report on the implementation of the rule triggering system. With this proposal, it is also hoped to show how MT can benefit today from Sign Language processing.

show abstract

Section: Traditional and Present Methodsmentioning

confidence: 99%

A rule triggering system for automatic text-to-sign translation

Filhol

Hadjadj

Testu

2015

Univ Access Inf Soc

View full text Add to dashboard Cite

show abstract

“…These approaches, designed at the beginning for written or speech processing, have been applied with some adaptations in the field of SL processing [21][22][23]27].…”

Section: Kazoo Language Processing: Sl-specific Approachmentioning

confidence: 99%

KAZOO: a sign language generation platform based on production rules

Braffort

Filhol

Delorme

et al. 2015

Univ Access Inf Soc

View full text Add to dashboard Cite

This paper describes KAZOO, a web application for sign language (SL) generation using a virtual signer. Firstly, it explains the motivation to this project, which is grounded on an approach designed solely from SL corpus analysis and modelling. Then, various projects conducted in the past few years on linguistic modelling and 3D animation are presented. The platform's architecture integrates parts of this work and new pieces of software allowing control and linking of all these components. This is an ongoing project, though the current version offers the possibility to animate a virtual signer automatically and synthesize the contents using an abstract representation, the authors' own linguistic model AZee, providing a means of validating this model.

show abstract

“…Further, an Example Based MT (EBMT) approach for Irish SL was proposed in (Morrissey & Way 2005;Morissey 2008). Wu et al (2007) proposed a Chinese SL MT system that uses a hybrid transfer based statistical model for translating Chinese text to Taiwanese SL (TSL). None of the systems described above can be directly used for ISL MT because sign languages over the world are not universal.…”

Section: Related Workmentioning

confidence: 99%

A Framework for the Automatic Generation of Indian Sign Language

Dasgupta¹,

Basu²,

Bhowmick³

et al. 2010

Journal of Intelligent Systems

View full text Add to dashboard Cite

Sign languages are visual-spatial natural languages having their own set of vocabularies, syntax and grammatical structures. Yet, designing machine translation system between spoken and sign languages generates a number of interesting challenges, mainly due to the difference in structure as well as modality of the source and target language pairs. This paper presents a rule-based framework for the English-To-Indian Sign Language machine translation system. The system can be used to disseminate information to the deaf people in India. The present system takes an English sentence as input, performs syntactic analysis, and represents the source language into an intermediate case frame representation. Finally, transfer grammar rules are used to generate the corresponding ISL sentence structure. As ISL does not have any written form, the final sentence is synthesized in terms of pre-recorded video streams. The prototype system has been primarily evaluated by the native signers of India. Our initial evaluation result shows a significant improvement over the existing baseline system.

show abstract

Transfer-based statistical translation of Taiwanese sign language using PCFG

Cited by 17 publications

References 16 publications

A rule triggering system for automatic text-to-sign translation

A rule triggering system for automatic text-to-sign translation

KAZOO: a sign language generation platform based on production rules

A Framework for the Automatic Generation of Indian Sign Language

Contact Info

Product

Resources

About