Author Login IJIRT Partner Network (IPN) Login

Audio Classification of Cats and Dogs using Python

  • Unique Paper ID: 179134
  • Volume: 11
  • Issue: 12
  • PageNo: 5366-5371
  • Abstract:
  • This paper presents a novel edge-optimized deep learning framework for real-time classification of cat and dog vocalizations, addressing key challenges in residential audio monitoring. Leveraging a MobileNetV2-inspired CNN architecture trained on MFCC features (Davis & Mermelstein, 1980), our solution achieves 94.32% accuracy (F1=0.93) while reducing model size by 43% compared to ResNet-18 baselines. The pipeline incorporates: • Robust preprocessing: Noise filtering + adaptive segmentation • Targeted augmentation: Time stretching (±20%) and pitch shifting (±2 semitones) • Edge deployment: <3s inference on Raspberry Pi (validated via stratified cross-validation) Outperforming SVM approaches by 12.7% (p<0.01), this work enables practical applications in smart pet care and veterinary acoustics. Future extensions will explore IoT integration and multi-species classification.
Download article email to a friend

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 12
  • PageNo: 5366-5371

Audio Classification of Cats and Dogs using Python

Related Articles

Impact Factor
8.01 (Year 2024)
UGC Approved
Journal no 47859

Join Our IPN

IJIRT Partner Network

Submit your research paper and those of your network (friends, colleagues, or peers) through your IPN account, and receive 800 INR for each paper that gets published.

Join Now

Latest Publication

Archive

View More

Recent Conferences

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024 Last Date: 15th March 2024

Submit inquiry