Whisper-Large-v3-Turbo Arabic (Fine-Tuned Standalone FP16)

Fine-tuned Whisper-Large-v3-Turbo for Arabic automatic speech recognition, delivering an optimal balance of accuracy (14.15% WER) and efficiency (809M parameters, 4 decoder layers).

• High-throughput inference: For 17.5x realtime deployment, see the INT8 quantized version: Omar10lfc/whisper-large-v3-turbo-arabic-ct2-int8.

Arabic Speech Recognition Benchmark (Common Voice 17.0)

All models were evaluated on the canonical Mozilla Common Voice 17.0 Arabic test set using the standardized normalization protocol:

Model Details

• Base Architecture: openai/whisper-large-v3-turbo (809M parameters, 32 encoder layers, 4 decoder layers)
• Training Method: Parameter-Efficient QLoRA ($r=64$, $\alpha=128$, 1,200 training steps) fused into unquantized FP16 weights.
• Language: Arabic (ar)

FP16 (Unquantized) vs. CTranslate2 INT8 Quantization Comparison

Does 8-bit quantization degrade accuracy? Our empirical benchmark across the full Mozilla Common Voice Arabic evaluation suite demonstrates quantization parity—the Word Error Rate (WER) difference between unquantized FP16 and CTranslate2 Static INT8 is statistically negligible ($\le 0.5$ to $0.8$ percentage points), while reducing disk size and VRAM by ~50% and accelerating inference by up to $3.6\times$:

Production Recommendations

• Deploy CTranslate2 INT8 (*-ct2-int8): Recommended for production APIs, serverless containers (Modal, RunPod, Hugging Face Spaces), and edge environments to cut cold-start load times and memory by half while maximizing throughput.
• Deploy PyTorch FP16: Recommended for further domain fine-tuning, research workflows, or systems requiring native Hugging Face transformers pipeline integration.

Quickstart (Transformers)

import torch
from transformers import AutoModelForConditionalGeneration, AutoProcessor, pipeline

model_id = "Omar10lfc/whisper-large-v3-turbo-arabic"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForConditionalGeneration.from_pretrained(
    model_id, 
    torch_dtype=torch.float16, 
    device_map="cuda"
)

pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    generate_kwargs={"language": "arabic", "task": "transcribe"}
)

result = pipe("path/to/arabic_audio.mp3")
print("Transcription:", result["text"])