基于参数迁移和卷积循环神经网络的语音情感识别

在语音情感识别研究中,已有基于深度学习的方法大多没有针对语音时频两域的特征进行建模,且存在网络模型训练时间长、识别准确性不高等问题。语谱图是语音信号转换后具有时频两域的特殊图像,为了充分提取语谱图时频两域的情感特征,提出了一种基于参数迁移和卷积循环神经网络的语音情感识别模型。该模型把语谱图作为网络的输入,引入AlexNet网络模型并迁移其预训练的卷积层权重参数,将卷积神经网络输出的特征图重构后输入LSTM（Long Short-Term Memory）网络进行训练。实验结果表明,所提方法加快了网络训练的速度,并提高了情感识别的准确率。     

基于决策边界优化域自适应的跨库语音情感识别

域自适应算法被广泛应用于跨库语音情感识别中;然而,许多域自适应算法在追求减小域差异的同时,丧失了目标域样本的鉴别性,导致其以高密度的形式存在于模型决策边界处,降低了模型的性能。基于此,提出一种基于决策边界优化域自适应（DBODA）的跨库语音情感识别方法。首先利用卷积神经网络进行特征处理,随后将特征送入最大化核范数及均值差异（MNMD）模块,在减小域间差异的同时,最大化目标域情感预测概率矩阵的核范数,从而提升目标域样本的鉴别性并优化决策边界。在以Berlin、eNTERFACE和CASIA语音库为基准库设立的六组跨库实验中,所提方法的平均识别精度领先于其他算法1.68～11.01个百分点,说明所提模型有效降低了决策边界的样本密度,提升了预测的准确性。     

Deep and shallow features fusion based on deep convolutional neural network for speech emotion recognition

Recent years have witnessed the great progress for speech emotion recognition using deep convolutional neural networks (DCNNs). In order to improve the performance of speech emotion recognition, a novel feature fusion method is proposed. With going deeper of the convolutional layers, the convolutional feature of traditional DCNNs gradually become more abstract, which may not be the best feature for speech emotion recognition. On the other hand, the shallow feature includes only global information without the detailed information extracted by deeper convolutional layers. According to these observations, we design a deep and shallow feature fusion convolutional network, which combines the feature from different levels of network for speech emotion recognition. The proposed network allows us to fully exploit deep and shallow feature. The popular Berlin data set is used in our experiments, the experimental results show that our proposed network can further improve speech emotion recognition rate which demonstrates the effectiveness of the proposed network.     

基于伪标签深度学习的高光谱影像半监督分类

研究了一种仅利用少量标记点训练深度卷积神经网络并对高光谱影像进行分类的方法。以图像分割获得的同质区增加训练样本数目；借助这些增加的样本训练初始分类器并预测所有未知点的初始类别；将每一初始类别聚集为适当的类簇，以类簇号作为伪标签对深度卷积网进行预训练；最后利用经过同质区增加的训练样本精调预训练深度卷积网。实验结果证明新方法可以在仅用少量实际标记样本的情况下成功地训练深度卷积网，对高光谱数据进行有效分类。     

Emotion Recognition with Capsule Neural Network

For human-machine communication to be as effective as human-to-human communication, research on speech emotion recognition is essential. Among the models and the classifiers used to recognize emotions, neural networks appear to be promising due to the network’s ability to learn and the diversity in configuration. Following the convolutional neural network, a capsule neural network (CapsNet) with inputs and outputs that are not scalar quantities but vectors allows the network to determine the part-whole relationships that are specific 6 for an object. This paper performs speech emotion recognition based on CapsNet. The corpora for speech emotion recognition have been augmented by adding white noise and changing voices. The feature parameters of the recognition system input are mel spectrum images along with the characteristics of the sound source, vocal tract and prosody. For the German emotional corpus EMO-DB, the average accuracy score for 4 emotions, neutral, boredom, anger and happiness, is 99.69%. For Vietnamese emotional corpus BKEmo, this score is 94.23% for 4 emotions, neutral, sadness, anger and happiness. The accuracy score is highest when combining all the above feature parameters, and this score increases significantly when combining mel spectrum images with the features directly related to the fundamental frequency.     

深度卷积神经网络的多GPU并行框架

近年来,深度卷积神经网络在图像识别和语音识别等领域被广泛运用,取得了很好的效果。深度卷积神经网络是层数较多的卷积神经网络,有数千万参数需要学习,计算开销大,导致训练非常耗时。针对这种情况,本文提出深度卷积神经网络的多GPU并行框架,设计并实现模型并行引擎,依托多GPU的强大协同并行计算能力,结合深度卷积神经网络在训练中的并行特点,实现快速高效的深度卷积神经网络训练。      

基于改进的卷积神经网络的中文情感分类

探究了基于卷积神经网络的句子级别的中文文本情感分类,模型以文本经过预处理后得到的词向量作为输入。传统的卷积神经网络是由线性卷积层、池化层和全连接层堆叠起来的,提出以跨通道卷积层替代传统线性卷积滤波器,对基本的卷积神经网络进行改进,提高网络的表达能力。实验表明,改进后的卷积神经网络在保证训练速度的情况下,识别率达到91.89%,优于传统的卷积神经网络,有较好的识别能力。     

基于混合分布注意力机制与混合神经网络的语音情绪识别方法

针对现有语音情绪识别中存在无关特征多和准确率较差的问题,提出一种基于混合分布注意力机制与混合神经网络的语音情绪识别方法。该方法在2个通道内,分别使用卷积神经网络和双向长短时记忆网络进行语音的空间特征和时序特征提取,然后将2个网络的输出同时作为多头注意力机制的输入矩阵。同时,考虑到现有多头注意力机制存在的低秩分布问题,在注意力机制计算方式上进行改进,将低秩分布与2个神经网络的输出特征的相似性做混合分布叠加,再经过归一化操作后将所有子空间结果进行拼接,最后经过全连接层进行分类输出。实验结果表明,基于混合分布注意力机制与混合神经网络的语音情绪识别方法比现有其他方法的准确率更高,验证了所提方法的有效性。     

基于自动弱标注数据的跨领域命名实体识别

近年来,在大规模标注语料上训练的神经网络模型大大提升了命名实体识别任务的性能.但是,新领域人工标注数据获取代价高昂,如何快速、低成本地进行领域迁移就显得非常重要.在目标领域仅给定无标注数据的情况下,该文尝试自动构建目标领域的弱标注语料并对其建模.首先,采用两种不同的方法对无标注数据进行自动标注;然后,采用留"同"去"异...     

结合数据平衡和注意力机制的CNN+LSTM的自然语音情感识别

为了解决语音情感识别中数据集样本分布不平衡的问题,提出一种结合数据平衡和注意力机制的卷积神经网络(CNN)和长短时记忆单元(LSTM)的语音情感识别方法.该方法首先对语音情感数据集中的语音样本提取对数梅尔频谱图,并根据样本分布特点对进行分段处理,以便实现数据平衡处理,通过在分段的梅尔频谱数据集中微调预训练好的CNN模型,用于学习高层次的片段语音特征.随后,考虑到语音中不同片段区域在情感识别作用的差异性,将学习到的分段CNN特征输入到带有注意力机制的LSTM中,用于学习判别性特征,并结合LSTM和Softmax层从而实现语音情感的分类.在BAUM-1s和CHEAVD2.0数据集中的实验结果表明,本文提出的语音情感识别方法能有效地提高语音情感识别性能.     

Emotion recognition from speech using sub-syllabic and pitch synchronous spectral features

In this work, spectral features extracted from sub-syllabic regions and pitch synchronous analysis are proposed for speech emotion recognition. Linear prediction cepstral coefficients, mel frequency cepstral coefficients and the features extracted from high amplitude regions of spectrum are used to represent emotion specific spectral information. These features are extracted from consonant, vowel and transition regions of each syllable to study the contribution of these regions toward recognition of emotions. Consonant, vowel and the transition regions are determined using vowel onset points. Spectral features extracted from each pitch cycle, are also used to recognize emotions present in speech. The emotions used in this study are: anger, fear, happy, neutral and sad. The emotion recognition performance using sub-syllabic speech segments are compared with the results of conventional block processing approach, where entire speech signal is processed frame by frame. The proposed emotion specific features are evaluated using simulated emotion speech corpus, IITKGP-SESC (Indian Institute of Technology, KharaGPur-Simulated Emotion Speech Corpus). The emotion recognition results obtained using IITKGP-SESC are compared with the results of Berlin emotion speech corpus. Emotion recognition systems are developed using Gaussian mixture models and auto-associative neural networks. The purpose of this study is to explore sub-syllabic regions to identify the emotions embedded in a speech signal, and if possible, to avoid processing of entire speech signal for emotion recognition without serious compromise in the performance.     

低信噪比下联合训练生成对抗网络的语音分离

提升低信噪比下的分离语音质量是语音分离技术研究的重点,而大多数语音分离方法在低信噪比下仍只对目标说话人的语音进行特征训练.针对目前方法的不足,提出了一种基于联合训练生成对抗网络GAN的混合语音分离方法.为避免复杂的声学特征提取,生成模型采用全卷积神经网络直接提取混合语音时域波形的高维特征,判别模型通过构建二分类卷积神经网络来学习干扰说话人的特征信息,继而使系统得到的分离信息来源不再单一.实验结果表明,所提方法在低信噪比下仍能更好地恢复高频成分的信息,在双说话人混合语音数据集上的分离性能要优于所对比的方法.     

Neural network-based symbol recognition using a few labeled samples

The recognition of pen-based visual patterns such as sketched symbols is amenable to supervised machine learning models such as neural networks. However, a sizable, labeled training corpus is often required to learn the high variations of freehand sketches. To circumvent the costs associated with creating a large training corpus, improve the recognition accuracy with only a limited amount of training samples and accelerate the development of sketch recognition system for novel sketch domains, we present a neural network training protocol that consists of three steps. First, a large pool of unlabeled, synthetic samples are generated from a small set of existing, labeled training samples. Then, a Deep Belief Network (DBN) is pre-trained with those synthetic, unlabeled samples. Finally, the pre-trained DBN is fine-tuned using the limited amount of labeled samples for classification. The training protocol is evaluated against supervised baseline approaches such as the nearest neighbor classifier and the neural network classifier. The benchmark data sets used are partitioned such that there are only a few labeled samples for training, yet a large number of labeled test cases featuring rich variations. Results suggest that our training protocol leads to a significant error reduction compared to the baseline approaches.     

Deep learning approaches for speech emotion recognition: state of the art and research challenges

Speech emotion recognition (SER) systems identify emotions from the human voice in the areas of smart healthcare, driving a vehicle, call centers, automatic translation systems, and human-machine interaction. In the classical SER process, discriminative acoustic feature extraction is the most important and challenging step because discriminative features influence the classifier performance and decrease the computational time. Nonetheless, current handcrafted acoustic features suffer from limited capability and accuracy in constructing a SER system for real-time implementation. Therefore, to overcome the limitations of handcrafted features, in recent years, variety of deep learning techniques have been proposed and employed for automatic feature extraction in the field of emotion prediction from speech signals. However, to the best of our knowledge, there is no in-depth review study is available that critically appraises and summarizes the existing deep learning techniques with their strengths and weaknesses for SER. Hence, this study aims to present a comprehensive review of deep learning techniques, uniqueness, benefits and their limitations for SER. Moreover, this review study also presents speech processing techniques, performance measures and publicly available emotional speech databases. Furthermore, this review also discusses the significance of the findings of the primary studies. Finally, it also presents open research issues and challenges that need significant research efforts and enhancements in the field of SER systems.

     

语音情感智能识别的建模与仿真

语音情感信息具有非线性、信息冗余、高维等复杂特点,数据含有大量噪声,传统识别模型难以消除冗余和噪声信息,导致语音情感识别正确率十分低.为了提高语音情感识别正确率,利用小波分析去噪和神经网络的非线性处理能力,提出一种基于过程神经元网络的语音情感智能识别模型.采用小波分析对语音情感信号进行去噪处理,利用主成分分析消除语音情感特征中的冗余信息,采用过程神经元网络对语音情感进行分类识别.仿真结果表明,基于过程神经元网络的识别模型的识别率比K近邻提高了13％,比支持向量机提高了8.75％,该模型是一种有效的语音情感智能识别工具.     

基于对抗训练策略的语言模型数据增强技术

基于最大似然估计（Maximum likelihood estimation,MLE）的语言模型（Language model,LM）数据增强方法由于存在暴露偏差问题而无法生成具有长时语义信息的采样数据.本文提出了一种基于对抗训练策略的语言模型数据增强的方法,通过一个辅助的卷积神经网络判别模型判断生成数据的真伪,从而引导递归神经网络生成模型学习真实数据的分布.语言模型的数据增强问题实质上是离散序列的生成问题.当生成模型的输出为离散值时,来自判别模型的误差无法通过反向传播算法回传到生成模型.为了解决此问题,本文将离散序列生成问题表示为强化学习问题,利用判别模型的输出作为奖励对生成模型进行优化,此外,由于判别模型只能对完整的生成序列进行评价,本文采用蒙特卡洛搜索算法对生成序列的中间状态进行评价.语音识别多候选重估实验表明,在有限文本数据条件下,随着训练数据量的增加,本文提出的方法可以进一步降低识别字错误率（Character error rate,CER）,且始终优于基于MLE的数据增强方法.当训练数据达到6M词规模时,本文提出的方法使THCHS30数据集的CER相对基线系统下降5.0%,AISHELL数据集的CER相对下降7.1%.     

Deep features-based speech emotion recognition for smart affective services

Emotion recognition from speech signals is an interesting research with several applications like smart healthcare, autonomous voice response systems, assessing situational seriousness by caller affective state analysis in emergency centers, and other smart affective services. In this paper, we present a study of speech emotion recognition based on the features extracted from spectrograms using a deep convolutional neural network (CNN) with rectangular kernels. Typically, CNNs have square shaped kernels and pooling operators at various layers, which are suited for 2D image data. However, in case of spectrograms, the information is encoded in a slightly different manner. Time is represented along the x-axis and y-axis shows frequency of the speech signal, whereas, the amplitude is indicated by the intensity value in the spectrogram at a particular position. To analyze speech through spectrograms, we propose rectangular kernels of varying shapes and sizes, along with max pooling in rectangular neighborhoods, to extract discriminative features. The proposed scheme effectively learns discriminative features from speech spectrograms and performs better than many state-of-the-art techniques when evaluated its performance on Emo-DB and Korean speech dataset.

     

基于深度残差网络的人脸表情识别

针对深度卷积神经网络随着卷积层数增加而导致网络模型难以训练和性能退化等问题,提出了一种基于深度残差网络的人脸表情识别方法。该方法利用残差学习单元来改善深度卷积神经网络模型训练寻优的过程,减少模型收敛的时间开销。此外,为了提高网络模型的泛化能力,从KDEF和CK+两种表情数据集上选取表情图像样本组成混合数据集用以训练网络。在混合数据集上采用十折（10-fold）交叉验证方法进行了实验,比较了不同深度的带有残差学习单元的残差网络与不带残差学习单元的常规卷积神经网络的表情识别准确率。当采用74层的深度残差网络时,可以获得90.79%的平均识别准确率。实验结果表明采用残差学习单元构建的深度残差网络可以解决网络深度和模型收敛性之间的矛盾,并能提升表情识别的准确率。     

改进的深度卷积网络及在碎纸片拼接中的应用

近年来,深度卷积网络在图像识别、语音识别和自然语言处理等领域广泛使用,取得了很好的效果。为解决全部样本均为无标签数据的分类问题,对深度卷积神经网络进行了改进,采用卷积自动编码器学习输入样本的特征,用k-均值聚类器代替深度卷积网络中的分类器,建立了改进的深度卷积网络结构,给出了相应的学习算法,将其用于解决碎纸片拼接问题。实验表明,该方法有效可行,提高了碎纸片拼接的准确性和鲁棒性。