Text to Sign Language Translation System: A Review of Literature

Text to Sign Language Translation System: A Review of Literature

Lalit Goyal (DAV College, Jalandhar, India) and Vishal Goyal (Punjabi University, Patiala, India)
Copyright: © 2016 |Pages: 16
DOI: 10.4018/IJSE.2016070104


Many machine translation systems for spoken languages are available, but the translation system between the spoken and Sign Language are limited. The translation from Text to Sign Language is different from the translation between spoken languages because the Sign Language is visual spatial language which uses hands, arms, face, and head and body postures for communication in three dimensions. The translation from text to Sign Language is complex as the grammar rules for Sign Language are not standardized. Still a number of approaches have been used for translating the Text to Sign Language in which the input is the text and output is in the form of pre-recorded videos or the animated character generated by computer (Avatar). This paper reviews the research carried out for automatic translation from Text to the Sign Language.
Article Preview


There are approximately 7105 known living languages in the world divided in 136 different language families. Among these 136 language families, Sign language is one and this family contains 136 sign languages all over the world depending upon the region of the world. Sign language is used by hearing impaired people to convey their message (http://www.wfdeaf.org).

Sign Language (SL) is a visual-spatial language which is used to communicate using hands, arms, face, head, and body postures. The signer uses the three dimensional space around his body to describe an event (Zeshan, 2003). Signs are categorized as manual signs (single handed or double handed) and non manual signs or combination of both as shown in Figure 1:

Figure 1.

ISL Type Hierarchy


Manual signs are those signs which uses the hand shapes, hand location, and hand movement where as non- manual component is the face expressions, head and body postures. A sign may have only manual part or only non- manual part or combination of both. For example, the sign “Yes” is signed by vertical head nod and it is pure non-manual component. One Handed Signs are represented by a single dominating hand where as two handed signs are represented by using both the hands. These signs can be either static or dynamic (having movements) and may include the non-manual component also. Figure 2 and Figure 3 show some examples of this categorization.

Figure 2.

One Handed Static Manual Sign (Ear) and Non-manual Sign (Headache)

Figure 3.

Two Handed Sign "Long" (Both hands are moving) and Sign "Flag" (Only the dominant right hand is moving)


Until the year 1960s, Sign languages were not viewed as bona fide languages, these were treated as the collection of gestures and mime. Dr. Stokoe’s research on American Sign Language proved that it is a full-fledged language with its own grammar, syntax, and other linguistic attributes. To prove the same for other sign languages, there are some efforts including Indian Sign Language (Zeshan, Vasishta, & Sethna, 2005).

This is worth mentioning here that sign languages are not “Natural languages represented through signs” or not even is word to word translation. This is because a particular word in spoken language can have multiple meanings and these multiple meanings cannot be represented by same sign. For example, the word ‘light’ in English has different meanings. Light means ‘not heavy’, or we say ‘light colour’ i.e. not dark or switch on the ‘light’. These different meanings in English are represented with same word ‘light’ but in sign language, all these meanings are represented by different signs. So, Sign Language is not representation of word as it is but rather the meanings of the words are represented using Sign Language.

Complete Article List

Search this Journal:
Open Access Articles
Volume 12: 2 Issues (2021): Forthcoming, Available for Pre-Order
Volume 11: 2 Issues (2020): 1 Released, 1 Forthcoming
Volume 10: 2 Issues (2019)
Volume 9: 2 Issues (2018)
Volume 8: 2 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2015)
Volume 5: 2 Issues (2014)
Volume 4: 2 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing