Machine Translation Systems for Indian Languages

Nair, Latha R; Peter, S. David

doi:10.5120/4785-7014

Cited by 21 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine Translation systems are either Rule Based or Statistical [8]. Statistical MT systems are dependent on the availability and quality of a large parallel corpus [9][10] [11] that is a good representative of the language and other lexical resources like Dictionary [1] [12]. During translation, a word is replaced with its possible translation from a dictionary obtained during training of the system or through external dictionary.…”

Section: Machine Translationmentioning

confidence: 99%

A Survey Report on the Existing Methods of Building a Parallel Corpus

Khosla¹

2018

ijarcs

View full text Add to dashboard Cite

This paper is a survey of the existing methods of building a parallel Corpus. The paper starts with a short introduction to a parallel corpus followed by the applications of a parallel corpus. Parallel corpus built in different language pairs and the method adopted is discussed and presented. The paper covers some of the methodologies of the major parallel corpus built. The survey report is restricted to corpus built aligned at sentence and document level.

show abstract

Section: Machine Translationmentioning

confidence: 99%

A Survey Report on the Existing Methods of Building a Parallel Corpus

Khosla¹

2018

ijarcs

View full text Add to dashboard Cite

show abstract

“…A combination of machine translations shows that the hybrid system is better than stand-alone systems: RBMT and EBMT [16], RBMT-SMT [19][20][21][22]. The hybrid Javanese machine translation is designed based on the corpus (Fig.…”

Section: The Design Of Javanese Translatormentioning

confidence: 99%

Hybrid machine translation for Javanese speech levels

Wibawa

Nafalski

Tweedale

et al. 2013

2013 5th International Conference on Knowledge and Smart Technology (KST)

View full text Add to dashboard Cite

Javanese is a local language with the biggest number of speakers in Indonesia. However, the fact that it is characterized by a complex system of politeness means that it is perceived negatively by Javanese teenagers, who have difficulty understanding it. The hybrid corpus-based machine translation described here is designed to address this by offering a means of translating speech levels appropriately. The system embeds statistical features into a memory-based machine translation to obtain the best performance of Javanese speech levels' translation. The evaluation shows satisfactory results; 0.83 and 90.4 for the average accuracy and quality of the translation, respectively.

show abstract

“…The dictionary-based approach relies on dictionary entries and word by word translations. These approaches require complete language grammatical structure data for translation from both ends which also consumes more time on complex datasets [8]. Due to this, the corpus-based approach is reliable in automatic translation process.…”

Section: Introductionmentioning

confidence: 99%

Lexical Based Reordering Models for English to Telugu Machine Translation

Vamsi,

Al Bataineh,

Doppala

2023

RIA

View full text Add to dashboard Cite

Telugu is one of the commonly spoken regional language in India. It is mostly spoken among the states of Andhra Pradesh and Telangana. In rural areas it is difficult for the people to understand the non-regional language specially while at the time of government works, land dealing transactions etc. Due to this there is a scope to develop a machine translation model from English to Telugu. The machine translation is an automatic technique of translating one language to another through Language Processing approach. To understand the Telugu language translation, the structural comparisons are done among English and Telugu languages to attain standard outcome. In this work, Lexical based reordering statistical model (LBRSM) is used for language conversion. This analyzes the language structure outcomes between word, phrase and hierarchical based models for the translation quality purpose. To maintain good quality translation TER and BLEU metrics 62.01 and 29.07 are considered for finding n-gram exact matches. From this work, the Phrase based reordering statistical model (PBRSM) achieved better results when compared with other system models in both training and testing phases.

show abstract

Machine Translation Systems for Indian Languages

Cited by 21 publications

References 9 publications

A Survey Report on the Existing Methods of Building a Parallel Corpus

A Survey Report on the Existing Methods of Building a Parallel Corpus

Hybrid machine translation for Javanese speech levels

Lexical Based Reordering Models for English to Telugu Machine Translation

Contact Info

Product

Resources

About