# Improving Cross-Corpus Speech Emotion Recognition with Adversarial Discriminative Domain Generalization (ADDoG)

arXiv: https://arxiv.org/abs/1903.12094 · HTML: https://mlsanity.com/paper/1903.12094/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Mar 2019.

## Built on by (most cited first, 19 of 19)

- [Stable Learning via Sparse Variable Independence](https://mlsanity.com/paper/2212.00992/) (2022)
- [Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentation](https://mlsanity.com/paper/2108.02510/) (2021)
- [Multimodal Emotion Recognition using Transfer Learning from Speaker Recognition and BERT-based models](https://mlsanity.com/paper/2202.08974/) (2022)
- [Robust Domain-Free Domain Generalization with Class-aware Alignment](https://mlsanity.com/paper/2102.08897/) (2021)
- [Few-shot Learning in Emotion Recognition of Spontaneous Speech Using a Siamese Neural Network with Adaptive Sample Pair Formation](https://mlsanity.com/paper/2109.02915/) (2021)
- [Cross-Corpora Spoken Language Identification with Domain Diversification and Generalization](https://mlsanity.com/paper/2302.05110/) (2023)
- [Learning with Alignments: Tackling the Inter- and Intra-domain Shifts for Cross-multidomain Facial Expression Recognition](https://mlsanity.com/paper/2407.05688/) (2024)
- [More Similar than Dissimilar: Modeling Annotators for Cross-Corpus Speech Emotion Recognition](https://mlsanity.com/paper/2509.12295/) (2025)
- [Acted vs. Improvised: Domain Adaptation for Elicitation Approaches in Audio-Visual Emotion Recognition](https://mlsanity.com/paper/2104.01978/) (2021)
- [A Comprehensive Review of Machine Learning Advances on Data Change: A Cross-Field Perspective](https://mlsanity.com/paper/2402.12627/) (2024)
- [Mouth Articulation-Based Anchoring for Improved Cross-Corpus Speech Emotion Recognition](https://mlsanity.com/paper/2412.19909/) (2024)
- [Unsupervised Pairwise Learning Optimization Framework for Cross-Corpus EEG-Based Emotion Recognition Based on Prototype Representation](https://mlsanity.com/paper/2508.11663/) (2025)
- [Accounting for Variations in Speech Emotion Recognition with Nonparametric Hierarchical Neural Network](https://mlsanity.com/paper/2109.04316/) (2021)
- [Domain Adapting Deep Reinforcement Learning for Real-world Speech Emotion Recognition](https://mlsanity.com/paper/2207.12248/) (2022)
- [Implicit Design Choices and Their Impact on Emotion Recognition Model Development and Evaluation](https://mlsanity.com/paper/2309.03238/) (2023)
- [Speaker Style-Aware Phoneme Anchoring for Improved Cross-Lingual Speech Emotion Recognition](https://mlsanity.com/paper/2509.20373/) (2025)
- [Test-Time Adaptation for Speech Emotion Recognition](https://mlsanity.com/paper/2601.16240/) (2026)
- [Boundary-aware Prototype-driven Adversarial Alignment for Cross-Corpus EEG Emotion Recognition](https://mlsanity.com/paper/2603.26713/) (2026)
- [Resource-Constrained Affect Modelling via Variance Regularisation Pruning](https://mlsanity.com/paper/2605.27479/) (2026)

## Abstract (arXiv)

Automatic speech emotion recognition provides computers with critical context to enable user understanding. While methods trained and tested within the same dataset have been shown successful, they often fail when applied to unseen datasets. To address this, recent work has focused on adversarial methods to find more generalized representations of emotional speech. However, many of these methods have issues converging, and only involve datasets collected in laboratory conditions. In this paper, we introduce Adversarial Discriminative Domain Generalization (ADDoG), which follows an easier to train "meet in the middle" approach. The model iteratively moves representations learned for each dataset closer to one another, improving cross-dataset generalization. We also introduce Multiclass ADDoG, or MADDoG, which is able to extend the proposed method to more than two datasets, simultaneously. Our results show consistent convergence for the introduced methods, with significantly improved results when not using labels from the target dataset. We also show how, in most cases, ADDoG and MADDoG can be used to improve upon baseline state-of-the-art methods when target dataset labels are added and in-the-wild data are considered. Even though our experiments focus on cross-corpus speech emotion, these methods could be used to remove unwanted factors of variation in other settings.