Scholarlink

<p>A&#xa0;technique for segmentation of speech signals into tonal components for monitoring postoperative voice dysfunction is proposed. The technique is based on the Teager energy operator, which increases sensitivity to amplitude and frequency changes in dysphonia. Validation using clinical data (1000 signals from 21&#xa0;patients) demonstrated low segmentation errors: 1.9% (Type&#xa0;I) and 1.7% (Type&#xa0;II), which is 3.1–21.3% more accurate than existing methods. This will enable effective identification of tonal components even when the amplitude of vocal cord vibration is reduced by 40–60%, enabling early diagnosis of recurrent laryngeal nerve palsy. The technique can be integrated into postoperative monitoring systems and opens up new opportunities for objective assessment of rehabilitation.</p>

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

A technique for segmentation of speech signals into tonal components for monitoring postoperative voice dysfunction

  • 期刊论文

  • 发表日期：

    2026年1月26日

查看全文
  • A. K. Alimuradov,
  • A. Yu. Tychkov,
  • Z. M. Yuldashev,
  • B. A. Porezanov,
  • A. A. Mamonova

摘要

A technique for segmentation of speech signals into tonal components for monitoring postoperative voice dysfunction is proposed. The technique is based on the Teager energy operator, which increases sensitivity to amplitude and frequency changes in dysphonia. Validation using clinical data (1000 signals from 21 patients) demonstrated low segmentation errors: 1.9% (Type I) and 1.7% (Type II), which is 3.1–21.3% more accurate than existing methods. This will enable effective identification of tonal components even when the amplitude of vocal cord vibration is reduced by 40–60%, enabling early diagnosis of recurrent laryngeal nerve palsy. The technique can be integrated into postoperative monitoring systems and opens up new opportunities for objective assessment of rehabilitation.