IndicTrans2: Towards High-Quality and Accessible Machine Translation Models for all 22 Scheduled Indian Languages.
AI4BharatJay P. GalaPranjal A. ChitaleRaghavan AKSumanth DoddapaneniVarun GummaAswanth KumarJanki NawaleAnupama SujathaRatish PuduppullyVivek RaghavanPratyush KumarMitesh M. KhapraRaj DabreAnoop KunchukuttanPublished in: CoRR (2023)
Keyphrases
- machine translation
- cross lingual
- language independent
- language processing
- natural language processing
- information extraction
- cross lingual information retrieval
- probabilistic model
- indian languages
- target language
- parallel corpus
- finite state transducers
- machine learning
- word alignment
- word level
- statistical machine translation
- translation model
- natural language generation
- cross language information retrieval
- word sense disambiguation
- artificial intelligence