在语音情感识别中搜索有效的预处理方法及基于 CNN 并采用高效通道注意力的架构
Searching for Effective Preprocessing Method and CNN-based Architecture with Efficient Channel Attention on Speech Emotion Recognition
AI总结：
针对语音情感识别数据不足与过拟合问题，本文搜索不同时频分辨率预处理方案，并提出带 ECA 的 6 层 CNN，在 IEMOCAP 上取得最高 80.28UA 和 80.46WA。
摘要：
语音情感识别（SER）利用计算机模型对语音中的人类情感进行分类。近年来，随着深度学习技术的应用，SER 的性能稳步提升。然而，与许多使用语音数据的领域不同，用于训练 SER 模型的数据并不充足。这会导致神经网络训练出现过拟合，进而造成性能下降。事实上，成功的情感识别需要一种有效的预处理方法，以及一种能够高效利用权重参数数量的模型结构。在本研究中，为搜索有效的情感语音预处理方法，我们提出使用八个具有不同频率-时间分辨率的数据集版本。我们提出一种带有高效通道注意力（ECA）的 6 层卷积神经网络（CNN）模型，以追求高效的模型结构。特别是，位置恰当的 ECA 模块仅需少量参数即可增强通道特征表示。在 interactive emotional dyadic motion capture（IEMOCAP）数据集上，提高情感语音预处理中的频率分辨率可以改善情感识别性能。此外，在深度卷积层之后使用 ECA 能够有效增强通道特征表示。因此，可获得最佳结果（79.37UA、79.68WA），超过以往 SER 模型的性能。此外，为弥补情感语音数据的不足，我们实验了多种预处理数据方法，这些方法通过一个样本生成采用所有不同设置预处理的可训练数据，从而实现数据增强。在实验中，我们取得了最高结果（80.28UA、80.46WA）。
英文摘要：
Speech emotion recognition (SER) classifies human emotions in speech with a computer model. Recently, performance in SER has steadily increased as deep learning techniques have adapted. However, unlike many domains that use speech data, data for training in the SER model is insufficient. This causes overfitting of training of the neural network, resulting in performance degradation. In fact, successful emotion recognition requires an effective preprocessing method and a model structure that efficiently uses the number of weight parameters. In this study, we propose using eight dataset versions with different frequency-time resolutions to search for an effective emotional speech preprocessing method. We propose a 6-layer convolutional neural network (CNN) model with efficient channel attention (ECA) to pursue an efficient model structure. In particular, the well-positioned ECA blocks can improve channel feature representation with only a few parameters. With the interactive emotional dyadic motion capture (IEMOCAP) dataset, increasing the frequency resolution in preprocessing emotional speech can improve emotion recognition performance. Also, ECA after the deep convolution layer can effectively increase channel feature representation. Consequently, the best result (79.37UA 79.68WA) can be obtained, exceeding the performance of previous SER models. Furthermore, to compensate for the lack of emotional speech data, we experiment with multiple preprocessing data methods that augment trainable data preprocessed with all different settings from one sample. In the experiment, we can achieve the highest result (80.28UA 80.46WA).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
