Codebase for Indic-Transliteration using Seq2Seq RNN. For latest repo with Transformer-based models, check: https://github.com/AI4Bharat/IndicXlit
Supported 14 Languages Bengali - বাংলা | Gujarati - ગુજરાતી | Hindi - हिंदी | Kannada - ಕನ್ನಡ | Konkani Goan - कोंकणी | Maithili - मैथिली | Malayalam - മലയാളം | Marathi - मराठी | Panjabi - ਪੰਜਾਬੀ | Sindhi - سنڌي | Sinhala - සිංහල | Telugu - తెలుగు | Tamil - தமிழ் | Urdu - اُردُو
Usage
apps/ai4bharat/transliteration/models
c. Export python path variable export PYTHONPATH=/path/to/transliteration
All the NN models (along with metadata) of Xlit - Transliteration are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Supported 14 Languages Bengali - বাংলা | Gujarati - ગુજરાતી | Hindi - हिंदी | Kannada - ಕನ್ನಡ | Konkani Goan - कोंकणी | Maithili - मैथिली | Malayalam - മലയാളം | Marathi - मराठी | Panjabi Eastern - ਪੰਜਾਬੀ | Sindhi - سنڌي | Sinhala - සිංහල | Telugu - తెలుగు | Tamil - தமிழ் | Urdu - اُردُو
Usage Download the xlit_apps_v0.4.1.zip file and follow the instructions in readme. (no need to clone the repo)
All the NN models (along with metadata) of Xlit - Transliteration are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Xlit - Transliteration Dataset by Story Weaver & AI4Bharat is licensed under a Creative Commons Attribution 4.0 International License.
Dataset Languages:
Check assets below.
Smaller models sizes and Vocab based reranking for better word prediction;
Updates for (1) Hindi (2) Konkani- Goan (3) Maithili
Neural Network based Back-transliteration for Indian languages, from English to Native Language
Neural Network based Transliteration for Indian languages, with low amount data. Languages supported: