# Cross-domain Speech Recognition with Unsupervised Character-level Distribution Matching

arXiv: https://arxiv.org/abs/2104.07491 · HTML: https://mlsanity.com/paper/2104.07491/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 29 references that are in the map.
- First posted on arXiv 15 Apr 2021.

## Builds on (most cited first, 14 of 14)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Deep Domain Confusion: Maximizing for Domain Invariance](https://mlsanity.com/paper/1412.3474/) (2014)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learning](https://mlsanity.com/paper/1609.06773/) (2016)
- [A Comparative Study on Transformer vs RNN in Speech Applications](https://mlsanity.com/paper/1909.06317/) (2019)
- [Transferrable Prototypical Networks for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1904.11227/) (2019)
- [CTC-Segmentation of Large Corpora for German End-to-end Speech Recognition](https://mlsanity.com/paper/2007.09127/) (2020)
- [Joint Geometrical and Statistical Alignment for Visual Domain Adaptation](https://mlsanity.com/paper/1705.05498/) (2017)
- [Domain Adversarial Training for Accented Speech Recognition](https://mlsanity.com/paper/1806.02786/) (2018)
- [Unsupervised Domain Adaptation for Speech Recognition via Uncertainty Driven Self-Training](https://mlsanity.com/paper/2011.13439/) (2020)
- [Transfer Learning from Adult to Children for Speech Recognition: Evaluation, Analysis and Recommendations](https://mlsanity.com/paper/1805.03322/) (2018)
- [Learning Noise-Invariant Representations for Robust Speech Recognition](https://mlsanity.com/paper/1807.06610/) (2018)
- [Libri-Adapt: A New Speech Dataset for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2009.02814/) (2020)

## Built on by (most cited first, 9 of 9)

- [Balancing Discriminability and Transferability for Source-Free Domain Adaptation](https://mlsanity.com/paper/2206.08009/) (2022)
- [Exploiting Adapters for Cross-lingual Low-resource Speech Recognition](https://mlsanity.com/paper/2105.11905/) (2021)
- [Boosting Cross-Domain Speech Recognition with Self-Supervision](https://mlsanity.com/paper/2206.09783/) (2022)
- [Advancing Test-Time Adaptation in Wild Acoustic Test Settings](https://mlsanity.com/paper/2310.09505/) (2023)
- [SAPIEN: Affective Virtual Agents Powered by Large Language Models](https://mlsanity.com/paper/2308.03022/) (2023)
- [Unsupervised domain adaptation for speech recognition with unsupervised error correction](https://mlsanity.com/paper/2209.12043/) (2022)
- [LI-TTA: Language Informed Test-Time Adaptation for Automatic Speech Recognition](https://mlsanity.com/paper/2408.05769/) (2024)
- [MADI: Inter-domain Matching and Intra-domain Discrimination for Cross-domain Speech Recognition](https://mlsanity.com/paper/2302.11224/) (2023)
- [Learning to Align: Addressing Character Frequency Distribution Shifts in Handwritten Text Recognition](https://mlsanity.com/paper/2506.09846/) (2025)

## Abstract (arXiv)

End-to-end automatic speech recognition (ASR) can achieve promising performance with large-scale training data. However, it is known that domain mismatch between training and testing data often leads to a degradation of recognition accuracy. In this work, we focus on the unsupervised domain adaptation for ASR and propose CMatch, a Character-level distribution matching method to perform fine-grained adaptation between each character in two domains. First, to obtain labels for the features belonging to each character, we achieve frame-level label assignment using the Connectionist Temporal Classification (CTC) pseudo labels. Then, we match the character-level distributions using Maximum Mean Discrepancy. We train our algorithm using the self-training technique. Experiments on the Libri-Adapt dataset show that our proposed approach achieves 14.39% and 16.50% relative Word Error Rate (WER) reduction on both cross-device and cross-environment ASR. We also comprehensively analyze the different strategies for frame-level label assignment and Transformer adaptations.