Unsupervised Domain Adaptation For Speech Recognition With Unsupervised Error Correction

Abstract

The transcription quality of automatic speech recognition (ASR) systems degrades significantly when transcribing audios coming from unseen domains. We propose an unsupervised error correction method for unsupervised ASR domain adaption, aiming to recover transcription errors caused by domain mismatch. Unlike existing correction methods that rely on transcribed audios for training, our approach requires only unlabeled data of the target domains in which a pseudo-labeling technique is applied to generate correction training samples. To reduce over-fitting to the pseudo data, we also propose an encoder-decoder correction model that can take into account additional information such as dialogue context and acoustic features. Experiment results show that our method obtains a significant word error rate (WER) reduction over non-adapted ASR systems. The correction model can also be applied on top of other adaptation approaches to bring an additional improvement of 10% relatively.

Stats

• citations4
• S2 citations—
• github stars0
• HF likes0
• heat score5.24
• arxiv keymai2022unsupervised

Related papers

• Unsupervised Domain Adaptation By Adversarial Learning For Robust Speech Recognition (2018)0.00
• Unsupervised Domain Adaptation For Robust Speech Recognition Via Variational Autoencoder-based Data Augmentation (2017)14.23
• Unsupervised Domain Adaptation For Speech Recognition Via Uncertainty Driven Self-training (2020)12.25
• Automatic Data Augmentation For Domain Adapted Fine-tuning Of Self-supervised Speech Representations (2023)0.00
• A Domain Adaptation Framework For Speech Recognition Systems With Only Synthetic Data (2025)5.24
• Boosting Cross-domain Speech Recognition With Self-supervision (2022)0.00
• Large-scale Domain Adaptation Via Teacher-student Learning (2017)13.93
• Cross-lingual Text-independent Speaker Verification Using Unsupervised Adversarial Discriminative Domain Adaptation (2019)11.85