Whisper Finetuned Amharic
W
Whisper Finetuned Amharic
Developed by seyyaw
Amharic speech recognition model fine-tuned from openai/whisper-small, achieving a word error rate of 2.0538% on the evaluation set
Downloads 57
Release Time : 2/1/2025
Model Overview
This model is an optimized automatic speech recognition (ASR) model for Amharic, suitable for Amharic speech-to-text tasks
Model Features
Low word error rate
Achieves a word error rate of 2.0538% on the evaluation set, demonstrating excellent performance
Efficient fine-tuning
Only requires 1 training epoch to achieve good results, with high training efficiency
Lightweight model
Based on whisper-small architecture, suitable for deployment in resource-constrained environments
Model Capabilities
Amharic speech recognition
Speech-to-text
Audio content transcription
Use Cases
Speech transcription
Amharic meeting minutes
Automatically convert Amharic meeting recordings into text transcripts
Highly accurate text transcription
Media content subtitle generation
Automatically generate subtitles for Amharic video content
Voice assistant
Amharic voice interaction
Supports Amharic voice command recognition
Featured Recommended AI Models
Š 2025AIbase