eess.ASJun 19, 2026

Towards Detecting Neural Audio Codec Synthesized Heart Sounds

Authors: Girish, Orchid Chetia Phukan, Mohd Mujtaba Akhtar, Bhavinkumar Vinodbhai Kuwar, Swarup Ranjan Behera, Arun Balaji Buduru

Organizations: UPES, India · NTHU, Taiwan · VBSPU, India · IIIT-Delhi, India · Independent Researcher, India

Abstract

In this paper, we introduce Synthetic Heart Sound Detection (SHAC), a task aimed at identifying phonocardiograms (PCGs) synthesized using neural audio codecs (NACs). To facilitate research in this direction, we release CARDIOFAKE, the first benchmark dataset for SHAC containing both real and codec-synthesized PCGs. We benchmark spectral representations (MFCC, LFCC) and self-supervised learning (SSL) representations (e.g., WavLM) for the task. Furthermore, we propose GROOT, a fusion framework that integrates spectral and SSL features for leveraging their complementary behavior. Experiments show that GROOT, combining MFCC and WavLM, achieves state-of-the-art performance, outperforming individual representations and competitive baselines.

Explore similar work

CardsList
  1. FlowFake: Liquid Networks for Audio Deepfake Detection

    Jun 17, 2026Shivaay Dhondiyal, Divyansh Sharma, Dinesh Kumar VishwakarmaAudio Deepfake DetectionFake News