Whisper Uz
W
Whisper Uz
Developed by jamshidahmadov
Uzbek speech recognition model fine-tuned on Whisper Base, trained on the Common Voice dataset
Downloads 1,179
Release Time : 12/16/2024
Model Overview
This model is an optimized speech recognition (ASR) model for the Uzbek language, fine-tuned based on OpenAI's Whisper Base architecture, suitable for Uzbek speech-to-text tasks
Model Features
Uzbek language optimization
Specially fine-tuned for Uzbek language, providing better speech recognition accuracy
Low word error rate
Achieved a word error rate (WER) of 14.0135 on the evaluation set, demonstrating excellent performance
Multi-dataset training
Trained on both Common Voice 17.0 and UzbekVoice datasets, covering a wider range of speech scenarios
Model Capabilities
Uzbek speech recognition
Speech-to-text
Automatic speech transcription
Use Cases
Speech transcription
Meeting minutes
Automatically convert Uzbek meeting recordings into text transcripts
14.01% word error rate
Media subtitle generation
Automatically generate subtitles for Uzbek video content
Voice assistant
Uzbek voice command recognition
Used to support smart voice assistants with Uzbek language capability
Featured Recommended AI Models
Š 2025AIbase