Open-source datasets (Samanantar) and models (IndicTrans) for neural machine translation between English and 12 Indic languages.

world

Open-source datasets and benchmarks (Aksharantar), models (IndicXlit), and applications for transliteration between Roman and scripts for 20+ Indic languages.

chart

Open-source models (IndicWav2Vec) for speech recognition in 9 Indian languages.

chat

Open-source language models (IndicBERT), benchmarks (IndicGLUE), and entity recognizers (IndicNER) for 10 Indian languages.

cloud

Open-source language generation model (IndicBART) and benchmarks (IndicNLG Suite) for 10 Indian languages.

shield

Open-source datasets (INCLUDE, SignCorpus) and models (OpenHands) for sign recognition for various 10 sign languages from around the world.

like

Open-source workbench for AI-assisted language work on Indian languages with initial focus on translation.

Open-source tool for AI-assisted video subtitling and translating with a focus on educational and media content.

code

AI4B

Lorem markdownum evinctus ut cape adhaeret gravis licet progenies ut haesit maxima ille. Est scorpius, mori vel in visaeque Haemoniis viperei furoris e ad vasti, distulit. Crudus sub coniuge iam: dea propera sive

Transliteration