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

Yorùbá language training text for NLP, ASR and TTS tasks

AfroLID, a powerful neural toolkit for African languages identification which covers 517 African languages.

Automatic Diacritic Restoration of Yorùbá language Text

Ìrànlọ́wọ́ is a utility library for analysis & (pre)processing of Yorùbá text → https://pypi.org/project/iranlowo

Website that hosts the African Voices projects. Users can download datasets and synthesizers, and synthesize speech in African languages

A Python based utility for Swahili and English number to words conversion.

Sankofa Display is a typeface that draws inspiration from African art styles, with a focus on straight-line geometric designs.

Code + data for the EMNLP'20 publication "Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages"

Lan_Tran is an app written in Kivy to make translations between Lantuosir and English easy! Lantuosir is a constructed language that I created based on Latin (and it's variants) & Bantu languages. It is developed as a fantasy lingua franca for the African Diaspora. The main influences are Spanish, English, and Yoruba.

Auto-generated stopwords for South African Bantu Languages

[morph] Scrape business stories to be used on TaxClock KE accessible at https://taxclock.codeforkenya.org/

Enterprise Python SDK for Abena AI Services - ASR, TTS, and Translation with multi-language support including African languages

Amharic Context Aware Spell Checker