MUHIGIRI Ashuza

@Ashuza11 · User

GitHub profile ↗ · Compare

Full-Stack AI Engineer | Audio AI, TTS, Music Generation, LLMs | Python, FastAPI, PyTorch | Building culturally grounded AI systems

@maarifa-co Kigali 13 followers94 repositories

Repositories

Ashuza11/ZeroMalaria

AI-assisted malaria triage, closed-loop referral & community intelligence for Rwanda’s CHWs

★ 0TypeScriptForks 0

Ashuza11/CongoLangBench

Curating text datasets for Congolese languages and benchmarking state-of-the-art language models across translation and language tasks.

★ 0HTMLForks 0

Ashuza11/Faida-app

A real-time business monitoring platform for airtime vendors

★ 0PythonForks 0

Ashuza11/SautiForge

An offline-first platform for collecting, transcribing, annotating, and managing multilingual speech datasets for low-resource languages.

★ 0TypeScriptForks 0

Ashuza11/Ashuza11

Ashuza11/Ashuza11 is a special repository: its README.md will appear on your profile.

★ 0Forks 0

Ashuza11/Ai-in-drCongo

Description: An evidence-based exploration of the Democratic Republic of the Congo’s emerging AI ecosystem, capabilities, challenges, and future.

★ 0PythonForks 0

Ashuza11/ntina-tarl

Code-switched voice agent that assesses children's reading level in their own language and runs level-appropriate practice, based on the Teaching at the Right Level (TaRL) method. Swahili–English and Yoruba–English. Built on Intron's Sahara voice APIs

★ 0JavaScriptForks 0

Ashuza11/faida-voice

Kinyarwanda voice ledger for informal vendors — sales, debts, savings, and a credit profile lenders can read.

★ 0TypeScriptForks 0

Ashuza11/sahara-codeswitch-agent

Code-switch aware voice agent for African telecom/fintech customer care, built on Intron's Sahara APIs — handles real Swahili/Kinyarwanda-English speech to resolve balance checks, failed transactions, and escalations end-to-end.

★ 0PythonForks 0

Ashuza11/google-waxal-asr

Multilingual Automatic Speech Recognition (ASR) system for African languages built for the Google WAXAL ASR Challenge on Zindi. Features fine-tuned speech models evaluated on WER and CER metrics.

★ 0PythonForks 0

Ashuza11/cassava-network-anomaly-ai

AI pipeline for telecom network anomaly detection, explanation, and action recommendations built for the Cassava AI Hackathon @ Deep Learning Indaba 2026.

★ 0Jupyter NotebookForks 0

Ashuza11/drc-bitext-eval

Parallel text collection and translation evaluation (BLEU / chrF++) for low-resource languages of the DRC.

★ 0Jupyter NotebookForks 0

Ashuza11/afrivoices-asr-hack

Multilingual ASR for Swahili, Kikuyu, Luo, Somali, Maasai, Kalenjin — AfriVoices East Africa Hackathon

★ 0PythonForks 0

Ashuza11/mashi-tts-bootstrap

Bootstrapping TTS for Mashi (Bashi/Shi): cross-lingual transfer from Kinyarwanda for an extremely low-resource Bantu language of eastern DRC — first Mashi speech dataset + dual-pipeline segmentation comparison.

★ 0PythonForks 0

Ashuza11/ntinaapp

Ntina — Audio-centric STEM knowledge base for Africa (PWA)

★ 0TypeScriptForks 0

Ashuza11/faida-landing

An intelligent management solution for telecom wholesalers in the DRC. Track inventory, debts, and sales in real time.

★ 0HTMLForks 0