# Revealing the Role of Audio Channels in ASR Performance Degradation

arXiv: https://arxiv.org/abs/2508.08967 · HTML: https://mlsanity.com/paper/2508.08967/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 12 Aug 2025.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Domain Generalization with MixStyle](https://mlsanity.com/paper/2104.02008/) (2021)
- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models](https://mlsanity.com/paper/2304.01933/) (2023)
- [VL-Adapter: Parameter-Efficient Transfer Learning for Vision-and-Language Tasks](https://mlsanity.com/paper/2112.06825/) (2021)
- [Google USM: Scaling Automatic Speech Recognition Beyond 100 Languages](https://mlsanity.com/paper/2303.01037/) (2023)
- [Real Time Speech Enhancement in the Waveform Domain](https://mlsanity.com/paper/2006.12847/) (2020)
- [BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition](https://mlsanity.com/paper/2109.13226/) (2021)
- [SpeechStew: Simply Mix All Available Speech Recognition Data to Train One Large Neural Network](https://mlsanity.com/paper/2104.02133/) (2021)
- [ASR Error Correction and Domain Adaptation Using Machine Translation](https://mlsanity.com/paper/2003.07692/) (2020)
- [CrisperWhisper: Accurate Timestamps on Verbatim Speech Transcriptions](https://mlsanity.com/paper/2408.16589/) (2024)
- [How Bad Are Artifacts?: Analyzing the Impact of Speech Enhancement Errors on ASR](https://mlsanity.com/paper/2201.06685/) (2022)
- [DeepFilterNet2: Towards Real-Time Speech Enhancement on Embedded Devices for Full-Band Audio](https://mlsanity.com/paper/2205.05474/) (2022)
- [Enhancing Pre-trained ASR System Fine-tuning for Dysarthric Speech Recognition using Adversarial Data Augmentation](https://mlsanity.com/paper/2401.00662/) (2024)
- [Noise Adaptive Speech Enhancement using Domain Adversarial Training](https://mlsanity.com/paper/1807.07501/) (2018)
- [How does end-to-end speech recognition training impact speech enhancement artifacts?](https://mlsanity.com/paper/2311.11599/) (2023)
- [Rethinking Processing Distortions: Disentangling the Impact of Speech Enhancement Errors on Speech Recognition Performance](https://mlsanity.com/paper/2404.14860/) (2024)
- [Exploiting Pre-Trained ASR Models for Alzheimer's Disease Recognition Through Spontaneous Speech](https://mlsanity.com/paper/2110.01493/) (2021)
- [Channel-Aware Domain-Adaptive Generative Adversarial Network for Robust Speech Recognition](https://mlsanity.com/paper/2409.12386/) (2024)
- [Naaloss: Rethinking the objective of speech enhancement](https://mlsanity.com/paper/2308.12615/) (2023)

## Abstract (arXiv)

Pre-trained automatic speech recognition (ASR) models have demonstrated strong performance on a variety of tasks. However, their performance can degrade substantially when the input audio comes from different recording channels. While previous studies have demonstrated this phenomenon, it is often attributed to the mismatch between training and testing corpora. This study argues that variations in speech characteristics caused by different recording channels can fundamentally harm ASR performance. To address this limitation, we propose a normalization technique designed to mitigate the impact of channel variation by aligning internal feature representations in the ASR model with those derived from a clean reference channel. This approach significantly improves ASR performance on previously unseen channels and languages, highlighting its ability to generalize across channel and language differences.