S

Sinai Voice Ar Stt

Developed by bakrianoo
An Arabic speech recognition model fine-tuned from facebook/wav2vec2-xls-r-300m on the Common Voice Arabic dataset
Downloads 29
Release Time : 3/2/2022

Model Overview

This is an Arabic Automatic Speech Recognition (ASR) model capable of converting Arabic speech into text. The model was fine-tuned on the Common Voice Arabic dataset and is suitable for standard Arabic speech recognition tasks.

Model Features

High-performance Arabic recognition
Achieved 18.1% Word Error Rate (WER) and 4.9% Character Error Rate (CER) on the Common Voice Arabic test set
Based on large-scale pretrained model
Fine-tuned from facebook/wav2vec2-xls-r-300m, inheriting powerful speech feature extraction capabilities
Supports language model-free inference
Can perform speech recognition directly without additional language model support

Model Capabilities

Arabic speech recognition
Speech-to-text
Automatic speech recognition

Use Cases

Speech transcription
Arabic speech transcription
Convert Arabic speech content into text
18.1% word error rate on standard Arabic test set
Voice assistants
Arabic voice command recognition
Used for voice command recognition in Arabic voice assistant systems
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase