Learning To Adapt: A Meta-learning Approach For Speaker Adaptation

Abstract

The performance of automatic speech recognition systems can be improved by adapting an acoustic model to compensate for the mismatch between training and testing conditions, for example by adapting to unseen speakers. The success of speaker adaptation methods relies on selecting weights that are suitable for adaptation and using good adaptation schedules to update these weights in order not to overfit to the adaptation data. In this paper we investigate a principled way of adapting all the weights of the acoustic model using a meta-learning. We show that the meta-learner can learn to perform supervised and unsupervised speaker adaptation and that it outperforms a strong baseline adapting LHUC parameters when adapting a DNN AM with 1.5M parameters. We also report initial experiments on adapting TDNN AMs, where the meta-learner achieves comparable performance with LHUC.

Stats

• citations19
• S2 citations—
• github stars0
• HF likes0
• heat score9.76
• arxiv keyklejch2018learning

Related papers

• Speaker Adaptive Training Using Model Agnostic Meta-learning (2019)9.92
• Learning Hidden Unit Contributions For Unsupervised Acoustic Model Adaptation (2016)14.47
• Unsupervised Model-based Speaker Adaptation Of End-to-end Lattice-free MMI Model For Speech Recognition (2022)2.26
• Bayesian Learning For Deep Neural Network Adaptation (2020)9.76
• Empirical Evaluation Of Speaker Adaptation On DNN Based Acoustic Model (2018)5.24
• Meta-tts: Meta-learning For Few-shot Speaker Adaptive Text-to-speech (2021)12.74
• Improved End-to-end Dysarthric Speech Recognition Via Meta-learning Based Model Re-initialization (2020)10.48
• A Unified Speaker Adaptation Method For Speech Synthesis Using Transcribed And Untranscribed Speech With Backpropagation (2019)0.00