Ganesh Sivaraman
Papers
1
Total Citations
8
H-Index
1
About
Ganesh Sivaraman’s research lies at the intersection of speech processing, deep learning, and audio forensics, with a particular focus on detecting manipulated or synthetic speech. His most-cited work, “Investigating voiced and unvoiced regions of speech for audio deepfake detection” (2025, 8 citations), addresses a critical gap in the field: while deep neural networks achieve high accuracy on benchmark datasets, they often lack interpretability, making it difficult for human evaluators to trust their decisions. Sivaraman’s contribution is to probe how different phonetic regions—voiced versus unvoiced—affect detection performance, offering a more transparent, explainable framework for identifying deepfakes. This work not only advances the technical frontier of audio forensics but also underscores the importance of building trustworthy AI systems. With a growing citation footprint and a commitment to bridging model accuracy with human interpretability, Sivaraman is emerging as a thoughtful voice in the fight against digital audio deception—a problem of increasing societal urgency.
Research Focus
Key Achievements
Top Papers
- 1