A simple, consistent and extendable toolkit for IndicTrans2. (Pypi: https://pypi.org/project/indictranstoolkit)
-
Updated
Apr 30, 2026 - Cython
#
indictrans2
Here are 16 public repositories matching this topic...
Layout-preserving PDF translator built for Indian scripts. HarfBuzz-shaped rendering keeps Kannada, Hindi, Tamil and 20+ languages pixel-perfect, with OCR for scans, legacy-font (Nudi/Kruti Dev) detection and Google or offline IndicTrans2 engines. Streamlit app, CLI and Python API.
multilingual python pdf unicode ocr translation tesseract hindi kannada harfbuzz indic-languages text-shaping indian-languages pymupdf document-translation streamlit pdf-translation indictrans2
-
Updated
Oct 4, 2026 - Python
desktop-app macos pdf ocr document-conversion agpl tauri paddleocr apple-silicon indic-nlp mineru vision-llm indictrans2
-
Updated
May 4, 2026 - TypeScript
AI-powered video translation pipeline that converts English-speaking videos into multiple Indic languages while preserving the speaker's voice and synchronizing realistic lip movements.
nlp translation computer-vision ffmpeg pytorch video-processing speech-recognition asyncio lip-sync whisper voice-cloning wav2lip generative-ai xtts multimodal-ai indictrans2
-
Updated
Jul 17, 2026 - Python
Multilingual concept prerequisite lookup across Indian school curriculum (NCERT) and Buddhist dharma terminology.
-
Updated
May 30, 2026 - Python
Open source multilingual voice OS — 22 Indian languages via AI4Bharat (IndicTrans2, IndicConformer, Parler-TTS, IndicXlit) + faster-whisper. Jarvis-style desktop assistant.
multilingual python open-source text-to-speech translation transliteration speech-recognition jarvis hindi tamil telugu whisper voice-assistant indian-languages fastapi indic-nlp faster-whisper ai4bharat indictrans2 parler-tts
-
Updated
Aug 20, 2026 - Python
A complete engineering case study in building an event-driven OCR pipeline: system design, distributed workers, hybrid AI model routing, and production observability - not just an OCR API wrapper.
ocr named-entity-recognition fasttext optical-character-recognition handwriting-recognition system-design event-driven-architecture document-processing fastapi redis-streams paddleocr document-intelligence structured-data-extraction qwen-vl indictrans2 hybrid-ocr multilingual-ocr indian-language-ocr regional-language-ocr
-
Updated
Sep 25, 2026 - Python
Engineering content translator — English ⇄ Marathi & Hindi. Built with IndicTrans2, FastAPI, Streamlit. BLEU 30.4 (Marathi), 42.6 (Hindi).
python nlp machine-translation transformers hindi marathi indian-languages fastapi huggingface streamlit ai4bharat indictrans2
-
Updated
May 12, 2026 - Python
A transliteration + translation pipeline benchmarking multiple NMT models for code-mixed Hinglish and Tanglish text
-
Updated
Jul 20, 2026 - Jupyter Notebook
An open-source conversational AI assistant for rural & semi-urban India. Supports voice-based queries in Hindi, Punjabi, Tamil, Bengali and more. Converts speech to text (ASR), retrieves answers via RAG, translates between Indic ↔ English, and responds back with Indic TTS. Built with AI4Bharat models, Whisper, Flan-T5, IndicTrans2, and IndicLID.
multilingual text-to-speech tts speech-to-text stt whisper conversational-ai speechrecognition rag voicebot digital-inclusion opensourceai llm flant5 ai4bharat rag-chatbot indiclanguages indictrans2 indiclid
-
Updated
Sep 8, 2025 - Python
Fine-tuning IndicTrans2 for Sanskrit-to-English translation on a low-resource parallel corpus, with BLEU/BERTScore evaluation and error analysis
nlp machine-translation transformers neural-machine-translation sanskrit bleu fine-tuning low-resource-languages huggingface indictrans2
-
Updated
Aug 8, 2026 - Jupyter Notebook
AI that reads tender documents and bids, checks every eligibility criterion, explains every verdict, and flags doubts for human review - making CRPF procurement faster, fairer, and fully auditable.
react redis opencv docker-compose postgresql tesseract spacy celery pymupdf fastapi streamlit muril sentiment-matching llama3 indictrans2
-
Updated
Jun 22, 2026 - Python
Offline multilingual AI speech translator for Indian languages using Whisper, IndicTrans2, and Raspberry Pi 4 with sub-10s edge inference.
multilingual python iot text-to-speech ai machine-translation speech transformers embedded-systems speech-to-text whisper nlp-machine-learning ras edge-ai offline-ai indictrans2
-
Updated
Jul 28, 2026 - Python
Android app for Hindi, English, and Marathi translation in all six directions, built with Kotlin, Jetpack Compose, FastAPI, and IndicTrans2.
-
Updated
Oct 7, 2026 - Kotlin
Book Translator is an ML-powered tool that translates books and documents from English to 20+ Indian languages using AI4Bharat's IndicTrans2 model, preserving PDF layout and formatting throughout the translation.
-
Updated
May 23, 2026 - Jupyter Notebook
AI-powered multilingual education platform for Jharkhand tribal schools, focused on Hindi–Santali translation, offline learning support, and foundational education.
android kotlin education offline-first machine-translation sih smart-india-hackathon fastapi santali indictrans2
-
Updated
Oct 5, 2026 - Kotlin
Add this topic to your repo
To associate your repository with the indictrans2 topic, visit your repo's landing page and select "manage topics."