# Enhanced ASR Robustness to Packet Loss with a Front-End Adaptation Network

arXiv: https://arxiv.org/abs/2406.18928 · HTML: https://mlsanity.com/paper/2406.18928/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 24 references that are in the map.
- First posted on arXiv 27 Jun 2024.

## Builds on (most cited first, 13 of 13)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [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)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech](https://mlsanity.com/paper/2205.12446/) (2022)
- [Can Contextual Biasing Remain Effective with Whisper and GPT-2?](https://mlsanity.com/paper/2306.01942/) (2023)
- [Deep speech inpainting of time-frequency masks](https://mlsanity.com/paper/1910.09058/) (2019)
- [INTERSPEECH 2022 Audio Deep Packet Loss Concealment Challenge](https://mlsanity.com/paper/2204.05222/) (2022)
- [Adversarial Auto-Encoding for Packet Loss Concealment](https://mlsanity.com/paper/2107.03100/) (2021)
- [ConcealNet: An End-to-end Neural Network for Packet Loss Concealment in Deep Speech Emotion Recognition](https://mlsanity.com/paper/2005.07777/) (2020)
- [tPLCnet: Real-time Deep Packet Loss Concealment in the Time Domain Using a Short Temporal Context](https://mlsanity.com/paper/2204.01300/) (2022)
- [Correcting Mispronunciations in Speech using Spectrogram Inpainting](https://mlsanity.com/paper/2204.03379/) (2022)

## Built on by (most cited first, 4 of 4)

- [ASR Under Noise: Exploring Robustness for Sundanese and Javanese](https://mlsanity.com/paper/2509.25878/) (2025)
- [On Inverse Problems, Parameter Estimation, and Domain Generalization](https://mlsanity.com/paper/2506.06024/) (2025)
- [Self-Supervised Test-Time Tuning for Packet Loss Concealment](https://mlsanity.com/paper/2607.01823/) (2026)
- [Parallel Time-Band Mixing with Learned Observation-Adding for Robust ASR Front-Ends](https://mlsanity.com/paper/2608.30326/) (2026)

## Abstract (arXiv)

In the realm of automatic speech recognition (ASR), robustness in noisy environments remains a significant challenge. Recent ASR models, such as Whisper, have shown promise, but their efficacy in noisy conditions can be further enhanced. This study is focused on recovering from packet loss to improve the word error rate (WER) of ASR models. We propose using a front-end adaptation network connected to a frozen ASR model. The adaptation network is trained to modify the corrupted input spectrum by minimizing the criteria of the ASR model in addition to an enhancement loss function. Our experiments demonstrate that the adaptation network, trained on Whisper's criteria, notably reduces word error rates across domains and languages in packet-loss scenarios. This improvement is achieved with minimal affect to Whisper model's foundational performance, underscoring our method's practicality and potential in enhancing ASR models in challenging acoustic environments.