DSTL: Solution to Limitation of Small Corpus in Speech Emotion Recognition

Traditional machine learning methods share a common hypothesis: training and testing datasets must be in a common feature space with the same distribution. However, in reality, the labeled target data may be rare, so that target space does not share the same feature space or distribution as an available training set (source domain). To address the mismatch of domains, we propose a Dual-Subspace Transfer Learning (DSTL) framework that considers both the common and specific information of the two domains. In DSTL, a latent common subspace is first learned to preserve the data properties and reduce the discrepancy of domains. Then, we propose a mapping strategy to transfer the sourcespecific information to the target subspace. The integration of the domain-common and specific information constructs the proposed DSTL framework. In comparison to the stateart-of works, the main contribution of our work is that the DSTL framework not only considers the commonalities, but also exploits the specific information. Experiments on three emotional speech corpora verify the effectiveness of our approach. The results show that the methods which include both domain-common and specific information perform better than the baseline methods which only exploit the domain commonalities.

• Publication year

  2019

• Venue

  Journal of Artificial Intelligence Research

• Publication date

  Unknown publication date

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1613/jair.1.11729
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Feature Selection Based Transfer Subspace Learning for Speech Emotion Recognition

  2020influential reference
• Transfer Linear Subspace Learning for Cross-Corpus Speech Emotion Recognition

  2019influential reference
• Speech Emotion Recognition Using Deep Convolutional Neural Network and Discriminant Temporal Pyramid Matching

  2018cited by this paper
• Semisupervised Autoencoders for Speech Emotion Recognition

  2018cited by this paper
• Speech Emotion Recognition with Emotion-Pair Based Framework Considering Emotion Distribution Information in Dimensional Emotion Space

  2017cited by this paper
• On the Behavior of Convolutional Nets for Feature Extraction

  2017cited by this paper
• Transfer learning for children's speech recognition

  2017cited by this paper
• Cross-corpus acoustic emotion recognition from singing and speaking: A multi-task learning approach

  2016cited by this paper
• Speech emotion recognition using transfer non-negative matrix factorization

  2016cited by this paper
• Cross-Corpus Speech Emotion Recognition Based on Domain-Adaptive Least-Squares Regression

  2016cited by this paper
• Discriminative Transfer Subspace Learning via Low-Rank and Sparse Representation

  2016cited by this paper
• Transfer learning for speech and language processing

  2015cited by this paper
• Detecting human emotion via speech recognition by using speech spectrogram

  2015cited by this paper
• A Primer on Neural Network Models for Natural Language Processing

  2015cited by this paper
• High-level feature representation using recurrent neural network for speech emotion recognition

  2015cited by this paper
• Annotations and consistency detection for Chinese dual-mode emotional speech database

  2015cited by this paper
• Information-Theoretic Multi-view Domain Adaptation: A Theoretical and Empirical Study

  2014cited by this paper
• Linked Source and Target Domain Subspace Feature Transfer Learning -- Exemplified by Speech Emotion Recognition

  2014cited by this paper
• Introducing shared-hidden-layer autoencoders for transfer learning and their application in acoustic emotion recognition

  2014influential reference
• Domain Adaptation for Face Recognition: Targetize Source Domain Bridged by Common Subspace

  2014cited by this paper
• Unsupervised Visual Domain Adaptation Using Subspace Alignment

  2013cited by this paper
• The INTERSPEECH 2013 computational paralinguistics challenge: social signals, conflict, emotion, autism

  2013cited by this paper
• Emotion recognition from speech: a review

  2012cited by this paper
• Emotion Recognition of Speech Using ANN and GMM

  2012cited by this paper
• Unsupervised learning in cross-corpus acoustic emotion recognition

  2011cited by this paper
• Emotion representation, analysis and synthesis in continuous space: A survey

  2011cited by this paper
• Cross-Corpus Acoustic Emotion Recognition: Variances and Strategies

  2010cited by this paper
• Bregman Divergence-Based Regularization for Transfer Subspace Learning

  2010cited by this paper
• A Survey on Transfer Learning

  2010influential reference
• Opensmile: the munich versatile and fast open-source audio feature extractor

  2010cited by this paper
• The INTERSPEECH 2009 emotion challenge

  2009cited by this paper
• Domain Adaptation via Transfer Component Analysis

  2009cited by this paper
• An Improved Hierarchical Multi-class Support Vector Machine with Binary Tree Architecture

  2008cited by this paper
• Transfer Learning via Dimensionality Reduction

  2008cited by this paper
• Domain Adaptation for Statistical Classifiers

  2006cited by this paper
• Analysis of Representations for Domain Adaptation

  2006cited by this paper
• Adaptive Emotion Recognition in Speech by Feature Selection Based on KL-divergence

  2006cited by this paper
• Graph embedding: a general framework for dimensionality reduction

  2005influential reference
• PERSONALITY PROCESSES AND INDIVIDUAL DIFFERENCES Universals and Cultural Differences in the Judgments of Facial Expressions of Emotion

  2004cited by this paper
• A Greedy EM Algorithm for Gaussian Mixture Learning

  2002cited by this paper
• In Defense of the Eight-Point Algorithm

  1997cited by this paper

• Design of Efficient Speech Emotion Recognition Based on Multi Task Learning

  2023influential citation
• Speech Emotion Recognition Based on Transfer Emotion-Discriminative Features Subspace Learning

  2023cites this paper
• Speech emotion recognition system based on wavelet transform and multi-task learning

  2022cites this paper
• Personalized Recommendation Algorithm of Literary Works Based on Annotated Corpus

  2022cites this paper
• Speech Emotion recognition feature Extraction and Classification

  2020cites this paper