# Emotion-Disentangled Embedding Alignment for Noise-Robust and Cross-Corpus Speech Emotion Recognition

arXiv: https://arxiv.org/abs/2510.09072 · HTML: https://mlsanity.com/paper/2510.09072/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 38 references that are in the map.
- First posted on arXiv 10 Oct 2025.

## Builds on (most cited first, 7 of 7)

- [Dawn of the transformer era in speech emotion recognition: closing the valence gap](https://mlsanity.com/paper/2203.07378/) (2022)
- [Domain Adversarial for Acoustic Emotion Recognition](https://mlsanity.com/paper/1804.07690/) (2018)
- [Contrastive Unsupervised Learning for Speech Emotion Recognition](https://mlsanity.com/paper/2102.06357/) (2021)
- [Multi-Task Semi-Supervised Adversarial Autoencoding for Speech Emotion Recognition](https://mlsanity.com/paper/1907.06078/) (2019)
- [A study on cross-corpus speech emotion recognition and data augmentation](https://mlsanity.com/paper/2201.03511/) (2022)
- [Affective social anthropomorphic intelligent system](https://mlsanity.com/paper/2304.11046/) (2023)
- [Self context-aware emotion perception on human-robot interaction](https://mlsanity.com/paper/2401.10946/) (2024)

## Abstract (arXiv)

Effectiveness of speech emotion recognition in real-world scenarios is often hindered by noisy environments and variability across datasets. This paper introduces a two-step approach to enhance the robustness and generalization of speech emotion recognition models through improved representation learning. First, our model employs EDRL (Emotion-Disentangled Representation Learning) to extract class-specific discriminative features while preserving shared similarities across emotion categories. Next, MEA (Multiblock Embedding Alignment) refines these representations by projecting them into a joint discriminative latent subspace that maximizes covariance with the original speech input. The learned EDRL-MEA embeddings are subsequently used to train an emotion classifier using clean samples from publicly available datasets, and are evaluated on unseen noisy and cross-corpus speech samples. Improved performance under these challenging conditions demonstrates the effectiveness of the proposed method.