whisper-tiny-informal-arabic Open Source Model - Free Implementation of Automatic Speech Recognition for Informal Arabic

Home

Whisper Tiny Informal Arabic

Developed by Raghad-DD

An automatic speech recognition model for informal Arabic, fine-tuned based on the Whisper Tiny architecture

Speech Recognition

Transformers

ArabicOpen Source License:Apache-2.0 #Arabic speech recognition #Informal speech processing #Lightweight model

Downloads 29

Release Time : 2/9/2025

Model Overview

This is an automatic speech recognition (ASR) model optimized for informal Arabic, fine-tuned on OpenAI's Whisper Tiny architecture, suitable for speech-to-text tasks in informal Arabic scenarios.

Model Features

Informal Arabic Optimization

Specially optimized for informal Arabic scenarios, better at recognizing everyday colloquial expressions

Lightweight Model

Based on the Whisper Tiny architecture, the model has a small footprint and is suitable for deployment in resource-limited environments

End-to-End Speech Recognition

Directly generates text output from raw audio input without complex preprocessing

Model Capabilities

Arabic speech recognition

Informal speech recognition

Real-time speech-to-text

Use Cases

Speech Transcription

Arabic Daily Conversation Transcription

Convert informal Arabic conversations into text records

Word error rate 63.7%

Customer Service Call Analysis

Analyze Arabic customer service call records

🚀 Whisper Tiny Informal Arabic

This model is a fine - tuned version of openai/whisper-tiny on the Informal Arabic dataset, aiming to provide high - quality automatic speech recognition for Arabic.

🚀 Quick Start

This model is a fine - tuned version of openai/whisper-tiny on the Informal Arabic dataset. It achieves the following results on the evaluation set:

Loss: 1.1448
Wer: 63.7075

✨ Features

Automatic Speech Recognition: Specialized for Informal Arabic speech recognition.
Fine - Tuned: Based on the openai/whisper - tiny model, fine - tuned on the Informal Arabic dataset.

🔧 Technical Details

Training and Evaluation Data

The model is trained on the Informal Arabic dataset (itskavya/gp). However, detailed information about the dataset is yet to be provided.

Training Procedure

Training Hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e - 05
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon = 1e - 08 and optimizer_args = No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 5000
mixed_precision_training: Native AMP

Training Results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5883	4.0650	1000	0.8899	66.7433
0.2235	8.1301	2000	0.9200	63.0635
0.0966	12.1951	3000	1.0207	63.3211
0.0389	16.2602	4000	1.1064	63.1555
0.0275	20.3252	5000	1.1448	63.7075

Framework Versions

Transformers 4.48.3
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0

📄 License

This model is licensed under the apache - 2.0 license.

📚 Documentation

Model Information

Property	Details
Library Name	transformers
Language	ar
Base Model	openai/whisper - tiny
Tags	automatic - speech - recognition, arabic, generated_from_trainer
Datasets	itskavya/gp
Metrics	wer
Model Index	Name: Whisper Tiny Informal Arabic, Results: Task - Automatic Speech Recognition on Informal Arabic dataset with Wer of 63.70745170193193

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご