Answers are grounded in this paper and cite their sources.
A robust and highly imperceptible audio watermarking technique is presented to secure the electronic patient record of Parkinson’s Disease (PD) affected patient. The proposed DCT-SVD based watermarking technique introduces minimal changes in speech such that the accuracy in classification of PD affected person’s speech and healthy person’s speech is retained. To achieve high imperceptibility the voiced part of the speech is considered for embedding the watermark. It is shown that the proposed watermarking technique is robust to common signal processing attacks. The practicability of the proposed technique is tested: by creating an android application to record & watermark the speech signal. The classification of PD affected speech is done using Support Vector Machine (SVM) classifier in cloud server.
No takes yet. Share an insight, caveat, or question.
Kanhe et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: