Scholarlink

This paper addresses the problem of embedding robust against AAC compression watermarks in audio files. We propose and analyze a method of embedding watermark in frequency domain (obtained after applying FFT) by using key-dependable dither modulation. This approach assures the same raw capacity but demonstrates better trade-of between audio quality and probability of errors than the embedding in the domain obtained after Haar wavelet transformation. The smaller probability of errors enables error correcting codes with higher rate to be used which results into bigger final capacity concerning information bits. We give some recommendations how to select proper embedding parameters such as quantizer step, which sub-bands to embed in as well as what error correcting codes to be used.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Watermarking Audio Content: Data Embedding Based on Fast Fourier Transform

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Todorka Alexandrova,
  • Hristo Kostadinov,
  • Nikolai L. Manev

摘要

This paper addresses the problem of embedding robust against AAC compression watermarks in audio files. We propose and analyze a method of embedding watermark in frequency domain (obtained after applying FFT) by using key-dependable dither modulation. This approach assures the same raw capacity but demonstrates better trade-of between audio quality and probability of errors than the embedding in the domain obtained after Haar wavelet transformation. The smaller probability of errors enables error correcting codes with higher rate to be used which results into bigger final capacity concerning information bits. We give some recommendations how to select proper embedding parameters such as quantizer step, which sub-bands to embed in as well as what error correcting codes to be used.