# A Curriculum Learning Method for Improved Noise Robustness in Automatic Speech Recognition

arXiv: https://arxiv.org/abs/1606.06864 · HTML: https://mlsanity.com/paper/1606.06864/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2016.

## Built on by (most cited first, 23 of 23)

- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [EfficientTrain: Exploring Generalized Curriculum Learning for Training Visual Backbones](https://mlsanity.com/paper/2211.09703/) (2022)
- [CL-MAE: Curriculum-Learned Masked Autoencoders](https://mlsanity.com/paper/2308.16572/) (2023)
- [EfficientTrain++: Generalized Curriculum Learning for Efficient Visual Backbone Training](https://mlsanity.com/paper/2405.08768/) (2024)
- [Data Augmentation in Training CNNs: Injecting Noise to Images](https://mlsanity.com/paper/2307.06855/) (2023)
- [Stochastic Re-weighted Gradient Descent via Distributionally Robust Optimization](https://mlsanity.com/paper/2306.09222/) (2023)
- [Selective Annotation via Data Allocation: These Data Should Be Triaged to Experts for Annotation Rather Than the Model](https://mlsanity.com/paper/2405.12081/) (2024)
- [Toddler-Guidance Learning: Impacts of Critical Period on Multimodal AI Agents](https://mlsanity.com/paper/2201.04990/) (2022)
- [Does the Definition of Difficulty Matter? Scoring Functions and their Role for Curriculum Learning](https://mlsanity.com/paper/2411.00973/) (2024)
- [Automatic Speech Recognition in the Modern Era: Architectures, Training, and Evaluation](https://mlsanity.com/paper/2510.12827/) (2025)
- [A bandit approach to curriculum generation for automatic speech recognition](https://mlsanity.com/paper/2102.03662/) (2021)
- [Noise2Recon: Enabling Joint MRI Reconstruction and Denoising with Semi-Supervised and Self-Supervised Learning](https://mlsanity.com/paper/2110.00075/) (2021)
- [Comparison and Analysis of New Curriculum Criteria for End-to-End ASR](https://mlsanity.com/paper/2208.05782/) (2022)
- [Selective Knowledge Distillation for Non-Autoregressive Neural Machine Translation](https://mlsanity.com/paper/2303.17910/) (2023)
- [Curriculum Guided Domain Adaptation in the Dark](https://mlsanity.com/paper/2308.00956/) (2023)
- [KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methods](https://mlsanity.com/paper/2308.11863/) (2023)
- [Curriculum optimization for low-resource speech recognition](https://mlsanity.com/paper/2202.08883/) (2022)
- [Robustness to Multi-Modal Environment Uncertainty in MARL using Curriculum Learning](https://mlsanity.com/paper/2310.08746/) (2023)
- [CHUCKLE -- When Humans Teach AI To Learn Emotions The Easy Way](https://mlsanity.com/paper/2510.09382/) (2025)
- [Analyzing Robustness of End-to-End Neural Models for Automatic Speech Recognition](https://mlsanity.com/paper/2208.08509/) (2022)
- [VALERIAN: Invariant Feature Learning for IMU Sensor-based Human Activity Recognition in the Wild](https://mlsanity.com/paper/2303.06048/) (2023)
- [SpacTor-T5: Pre-training T5 Models with Span Corruption and Replaced Token Detection](https://mlsanity.com/paper/2401.13160/) (2024)
- [End-to-end Jordanian dialect speech-to-text self-supervised learning framework](https://mlsanity.com/paper/2609.24410/) (2026)

## Abstract (arXiv)

The performance of automatic speech recognition systems under noisy environments still leaves room for improvement. Speech enhancement or feature enhancement techniques for increasing noise robustness of these systems usually add components to the recognition system that need careful optimization. In this work, we propose the use of a relatively simple curriculum training strategy called accordion annealing (ACCAN). It uses a multi-stage training schedule where samples at signal-to-noise ratio (SNR) values as low as 0dB are first added and samples at increasing higher SNR values are gradually added up to an SNR value of 50dB. We also use a method called per-epoch noise mixing (PEM) that generates noisy training samples online during training and thus enables dynamically changing the SNR of our training data. Both the ACCAN and the PEM methods are evaluated on a end-to-end speech recognition pipeline on the Wall Street Journal corpus. ACCAN decreases the average word error rate (WER) on the 20dB to -10dB SNR range by up to 31.4% when compared to a conventional multi-condition training method.