WMCodec: End-to-End Neural Speech Codec with Deep Watermarking for Authenticity Verification

TL;DR

WMCodec is an innovative neural speech codec that jointly trains for compression and watermarking, significantly improving authenticity verification robustness and imperceptibility in speech communication.

Contribution

It introduces the first end-to-end neural speech codec with integrated deep watermarking, enhancing verification accuracy and robustness over previous separate methods.

Findings

Outperforms AudioSeal with Encodec in quality metrics

Achieves over 99% watermark extraction accuracy under attacks

Maintains high imperceptibility and robustness at 6 kbps bandwidth

Abstract

Recent advances in speech spoofing necessitate stronger verification mechanisms in neural speech codecs to ensure authenticity. Current methods embed numerical watermarks before compression and extract them from reconstructed speech for verification, but face limitations such as separate training processes for the watermark and codec, and insufficient cross-modal information integration, leading to reduced watermark imperceptibility, extraction accuracy, and capacity. To address these issues, we propose WMCodec, the first neural speech codec to jointly train compression-reconstruction and watermark embedding-extraction in an end-to-end manner, optimizing both imperceptibility and extractability of the watermark. Furthermore, We design an iterative Attention Imprint Unit (AIU) for deeper feature integration of watermark and speech, reducing the impact of quantization noise on the watermark. Experimental results show WMCodec outperforms AudioSeal with Encodec in most quality metrics for watermark imperceptibility and consistently exceeds both AudioSeal with Encodec and reinforced TraceableSpeech in extraction accuracy of watermark. At bandwidth of 6 kbps with a watermark capacity of 16 bps, WMCodec maintains over 99% extraction accuracy under common attacks, demonstrating strong robustness.