One Shot Learning for Speech Separation
arXiv:2011.10233
Abstract
Despite the recent success of speech separation models, they fail to separate sources properly while facing different sets of people or noisy environments. To tackle this problem, we proposed to apply meta-learning to the speech separation task. We aimed to find a meta-initialization model, which can quickly adapt to new speakers by seeing only one mixture generated by those people. In this paper, we use model-agnostic meta-learning(MAML) algorithm and almost no inner loop(ANIL) algorithm in Conv-TasNet to achieve this goal. The experiment results show that our model can adapt not only to a new set of speakers but also noisy environments. Furthermore, we found out that the encoder and decoder serve as the feature-reuse layers, while the separator is the task-specific module.
arXiv commentAccepted to ICASSP 2021
References in corpus (9)
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
- Meta-Learning with Latent Embedding Optimization
- Voice Separation with an Unknown Number of Multiple Speakers
- Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML
- Meta-Learning for Low-Resource Neural Machine Translation
- Lifelong Domain Word Embedding via Meta-Learning
- Improving End-to-End Speech-to-Intent Classification with Reptile