Detection of AI-Synthesized Speech Using Cepstral & Bispectral Statistics
This addresses the need for digital audio forensics to combat speech clones and deepfakes, though it appears incremental as it builds on existing statistical analyses.
The paper tackles the problem of distinguishing human speech from AI-synthesized speech by proposing a method that integrates cepstral and bispectral analyses, achieving detection through a machine learning model.
Digital technology has made possible unimaginable applications come true. It seems exciting to have a handful of tools for easy editing and manipulation, but it raises alarming concerns that can propagate as speech clones, duplicates, or maybe deep fakes. Validating the authenticity of a speech is one of the primary problems of digital audio forensics. We propose an approach to distinguish human speech from AI synthesized speech exploiting the Bi-spectral and Cepstral analysis. Higher-order statistics have less correlation for human speech in comparison to a synthesized speech. Also, Cepstral analysis revealed a durable power component in human speech that is missing for a synthesized speech. We integrate both these analyses and propose a machine learning model to detect AI synthesized speech.