中文

ACM Multimedia 2023 计算副语言学挑战赛:情绪分享与请求

声音 2023-05-02 v2 机器学习 音频与语音处理

摘要

ACM Multimedia 2023 计算副语言学挑战赛(Computational Paralinguistics Challenge)首次在明确定义的条件下于研究竞赛中解决两个不同问题:在情绪分享子挑战(Emotion Share Sub-Challenge)中,需对语音进行回归;在请求子挑战(Requests Sub-Challenges)中,需检测请求与抱怨。我们描述了这些子挑战、基线特征提取,以及基于常用 ComPaRE 特征、auDeep 工具包和使用 DeepSpectRum 工具包从预训练 CNN 提取深度特征的分类器;此外,还使用了 wav2vec2 模型。

关键词

引用

@article{arxiv.2304.14882,
  title  = {The ACM Multimedia 2023 Computational Paralinguistics Challenge: Emotion Share & Requests},
  author = {Björn W. Schuller and Anton Batliner and Shahin Amiriparian and Alexander Barnhill and Maurice Gerczuk and Andreas Triantafyllopoulos and Alice Baird and Panagiotis Tzirakis and Chris Gagne and Alan S. Cowen and Nikola Lackovic and Marie-José Caraty and Claude Montacié},
  journal= {arXiv preprint arXiv:2304.14882},
  year   = {2023}
}

备注

5 pages, part of the ACM Multimedia 2023 Grand Challenge "The ACM Multimedia 2023 Computational Paralinguistics Challenge (ComParE 2023). arXiv admin note: text overlap with arXiv:2205.06799