Amharic Speech-to-Text Transcription
Group 10
Project Description
This project focuses on developing a system for transcribing Amharic speech into text. The system aims to provide accurate and efficient transcription capabilities for the Amharic language, leveraging state-of-the-art technologies in speech recognition and natural language processing.
Group Members
| Name | ID |
|---|---|
| Yosef Ayele Eshetu | UGR/2067/13 |
| Yonas Engdu | UGR/4575/13 |
| Yosef Aweke Dinku | UGR/5887/13 |
| Yosef Muluneh Bane | UGR/5715/13 |
Technologies
- Python for writing training scripts
- Facebook's Wav2Vec2 as the base model
- SpeechBrain for training
Datasets Used for Fine-tuning
facebook/2M-Belebelefsicoli/common_voice_19_0