amharic

Syllable-aware BPE tokenizer for the Amharic language (አማርኛ) – fast, accurate, trainable.

Amharic English Machine Translation Corpus prepared through website crawelling and custom preprocessing.

A Python package that can transliterate Latin characters to Geez characters and vice versa.

simple bs4 based web crawl for a corpus in need of statistical machine translation

Quran speech recognition (ASR) with verse matching, Iqra mode, and translations in Arabic, English, Somali, Amharic & Swahili. Built with Tarteel Whisper.

The set of files used for the development of the Amharic Corpus.

Amharic Large Language Model

Modular Amharic text preprocessing toolkit with composable processors and pipeline.

Amharic-Word Embedding-Word2vec is a pre-trained distributed word representation (word embedding) which aims to provide the Amharic NLP researcher with free to use.

This repository is an implementation of Amharic speech to text setup.

amharic search engine

Amharic Legal RAG + Sentiment Analysis pipeline for Ethiopian legal documents and complaints. Built for localized AI applications in Ethiopia.

A benchmark measuring AI agent reliability on paired Amharic and English tasks in simulated Ethiopian workflows.

Speech-to-text data collection with Kafka, Airflow, and Spark, building a pipeline that can be deployed to process posting and receiving text and audio files from and into a data lake, apply transformation in a distributed manner, and load it into a warehouse in a suitable format to train a speech-to-text model.

Amharic predictive text keyboard with next-word suggestion using N-gram and Bi-LSTM models. Built for faster Amharic typing.

Amharic Context Aware Spell Checker

የአማርኛ የልጆች ተረቶችን ወደ ድምጽ የሚቀይርና የሚያጫውት የ Streamlit መተግበሪያ።(A Streamlit application that converts and plays Amharic children's stories into audio.)

Generate & evaluate fine-tuning-ready African-language translation datasets, powered by Daraja AI's Babel models. Multi-signal eval (confidence, back-translation, LLM-as-judge) + SQLite persistence.

የአማርኛ የጥላቻ ንግግር መለያ እና የአገላለፅ ማስተካከያ መተግበሪያ ነው። ይህ ፕሮጀክት የማሽን መማሪያ ሞዴሎችን እና የጀሚኒ ኤፒአይን ይጠቀማል። (Amharic hate speech detection and polite rephrasing application. This project uses machine learning models and the Gemini API.)

ለኢትዮጵያ ገበሬዎች የሰብል ምርት ትንበያ፣ የማዳበሪያ ምክር፣ የምግብ ጥራት ምርመራ እና የአማርኛ AI አማካሪ የሚሰጥ መተግበሪያ ። (A platform providing crop yield prediction, fertilizer advice, food quality inspection, and an Amharic AI advisor for Ethiopian farmers.)