# DRAFT: A Novel Framework to Reduce Domain Shifting in Self-supervised Learning and Its Application to Children's ASR

arXiv: https://arxiv.org/abs/2206.07931 · HTML: https://mlsanity.com/paper/2206.07931/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jun 2022.

## Built on by (most cited first, 21 of 21)

- [Modular Deep Learning](https://mlsanity.com/paper/2302.11529/) (2023)
- [Benchmarking Children's ASR with Supervised and Self-supervised Speech Foundation Models](https://mlsanity.com/paper/2406.10507/) (2024)
- [Exploring Efficient-tuning Methods in Self-supervised Speech Models](https://mlsanity.com/paper/2210.06175/) (2022)
- [Adaptation of Whisper models to child speech recognition](https://mlsanity.com/paper/2307.13008/) (2023)
- [Improving Accented Speech Recognition with Multi-Domain Training](https://mlsanity.com/paper/2303.07924/) (2023)
- [A CTC Alignment-based Non-autoregressive Transformer for End-to-end Automatic Speech Recognition](https://mlsanity.com/paper/2304.07611/) (2023)
- [Children's Speech Recognition through Discrete Token Enhancement](https://mlsanity.com/paper/2406.13431/) (2024)
- [TICL+: A Case Study On Speech In-Context Learning for Children's Speech Recognition](https://mlsanity.com/paper/2512.18263/) (2025)
- [Some voices are too common: Building fair speech recognition systems using the Common Voice dataset](https://mlsanity.com/paper/2306.03773/) (2023)
- [Efficient Adapter Tuning of Pre-trained Speech Models for Automatic Speaker Verification](https://mlsanity.com/paper/2403.00293/) (2024)
- [Don't Stop Self-Supervision: Accent Adaptation of Speech Representations via Residual Adapters](https://mlsanity.com/paper/2307.00453/) (2023)
- [A comparative analysis between Conformer-Transducer, Whisper, and wav2vec2 for improving the child speech recognition](https://mlsanity.com/paper/2311.04936/) (2023)
- [Stable Distillation: Regularizing Continued Pre-training for Low-Resource Automatic Speech Recognition](https://mlsanity.com/paper/2312.12783/) (2023)
- [Robust Automatic Speech Recognition via WavAugment Guided Phoneme Adversarial Training](https://mlsanity.com/paper/2307.12498/) (2023)
- [UniPET-SPK: A Unified Framework for Parameter-Efficient Tuning of Pre-trained Speech Models for Robust Speaker Verification](https://mlsanity.com/paper/2501.16542/) (2025)
- [Can Layer-wise SSL Features Improve Zero-Shot ASR Performance for Children's Speech?](https://mlsanity.com/paper/2508.21225/) (2025)
- [Entropy-Aware Domain-Routed Mixture-of-Experts Speech-LLM Framework: A Case Study of Multi-Domain Child-Adult ASR](https://mlsanity.com/paper/2606.10454/) (2026)
- [Feature Normalization for Fine-tuning Self-Supervised Models in Speech Enhancement](https://mlsanity.com/paper/2306.08406/) (2023)
- [Examining Test-Time Adaptation for Personalized Child Speech Recognition](https://mlsanity.com/paper/2409.13095/) (2024)
- [G-IFT: A Gated Linear Unit adapter with Iterative Fine-Tuning for Low-Resource Children's Speaker Verification](https://mlsanity.com/paper/2508.07836/) (2025)
- [Cross-Dataset, Age, and Gender Generalization: A Comprehensive Analysis of Fine-Tuning Strategies for Low-Resource Children's ASR](https://mlsanity.com/paper/2606.19791/) (2026)

## Abstract (arXiv)

Self-supervised learning (SSL) in the pretraining stage using un-annotated speech data has been successful in low-resource automatic speech recognition (ASR) tasks. However, models trained through SSL are biased to the pretraining data which is usually different from the data used in finetuning tasks, causing a domain shifting problem, and thus resulting in limited knowledge transfer. We propose a novel framework, domain responsible adaptation and finetuning (DRAFT), to reduce domain shifting in pretrained speech models through an additional adaptation stage. In DRAFT, residual adapters (RAs) are inserted in the pretrained model to learn domain-related information with the same SSL loss as the pretraining stage. Only RA parameters are updated during the adaptation stage. DRAFT is agnostic to the type of SSL method used and is evaluated with three widely used approaches: APC, Wav2vec2.0, and HuBERT. On two child ASR tasks (OGI and MyST databases), using SSL models trained with un-annotated adult speech data (Librispeech), relative WER improvements of up to 19.7% are observed when compared to the pretrained models without adaptation. Additional experiments examined the potential of cross knowledge transfer between the two datasets and the results are promising, showing a broader usage of the proposed DRAFT framework.