Turi Abu
Graduate CS student
Tsinghua University
I am a Master's student in Computer Science at Tsinghua University, specializing in Speech Recognition. Currently, I am conducting research at the Center for Speech and Language Technologies (CSLT) under the supervision of Prof. Dong Wang and Prof. Thomas Fang Zheng. My research focuses on developing automatic speech recognition systems for the Oromo language, contributing to the advancement of low-resource language technologies. My research interests include ASR, TTS, and NLP.
News
December 21, 2024
🎉Paper accepted at ICASSP 2025 Conference
Our research paper 'Sagalee: An Open Source Automatic Speech Recognition Dataset for Oromo Langauge' has been accepted for presentation at the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025)
June 25, 2023
【毕业季】学校举行2023届优秀毕业生代表座谈会
School held the 2023 Symposium of Outstanding Graduate Representatives
December 6, 2022
Three International Students Awarded the Title: 2022 Excellent International Students of UESTC
Recipient of the 2022 Excellent International Students of UESTC award
Education
Tsinghua University
MSc in Computer Science
University of Electronic Science and Technology of China
BSc in Software Engineering
Publications
International Journal of Machine Learning and Cybernetics 2024
Unified Deep Learning Model for Multitask Representation and Transfer Learning: Image Classification, Object Detection, and Image Captioning
LY Bayisa, W Wang, Q Wang, CC Ukwuoma, Gutem HK, Endris A and Turi Abu
Journal of Public Health and Environmental Technology 2023
Attention-Based End-to-End Hybrid Ensemble Model for Breast Cancer Multi-Classification
CC Ukwuoma, D Cai, ES gATI, VK Agbesi, G Deribachew, LY Bayisa, and Turi Abu
Experience
Research Assitant — CSLT, Tsinghua University
Advisor: Prof Dong Wang, Prof Thomas Fang Zheng
Working on Automatic Speech Recognition for Oromo Language.
- Developed data collection mobile app.
- Created 100 hours of read speech dataset.
- Published experiment result and open sourced the dataset.
- Exploring different techniques to improve the ASR results, inluding Language modeling and finetuning strategies.
Intern-AI Engineer — HCYtech
Joined AI Department of the company. Key responsibilities included:
- Developed smoking detection system using YOLO, from surveillance camera feeds.
- Developed custom chatbot using OpenAI API and Speech-to-text using whisper model.
Portfolio
Mobile Application for Speech Data Collection
- Developed android application for Automatic Speech Recognition data colllection using Flutter and firebase
Local Retrieval Augmented Generation (RAG) Using Ollama
- Built a local RAG using open source llama model for question answering task
Finetuning LLM for Grammatical Error Correction (GEC)
- Fine-tuned T5 Flan-based model on JFLEG dataset for GEC task, and built Grammarly like website for demo
CodeDig - Semantic Code Search Engine
- Built a semantic code search engine using ChatGLM API, whoosh library, and Flask.
Distributed Database Management System
- Design and implementation of Distributed database system.