论文标题
一击学习以分离语音
One Shot Learning for Speech Separation
论文作者
论文摘要
尽管言语分离模型最近取得了成功，但它们在面对不同的人或嘈杂环境的同时未能正确分离来源。为了解决这个问题，我们建议将元学习应用于语音分离任务。我们的目的是找到一个元定位模型，该模型只能看到这些人产生的一种混合物，可以快速适应新的扬声器。在本文中，我们使用模型 - 静态元学习（MAML）算法，几乎没有Conv-Tasnet中的内部环（Anil）算法来实现此目标。实验结果表明，我们的模型不仅可以适应一组新的扬声器，还可以适应嘈杂的环境。此外，我们发现编码器和解码器用作功能固定层，而分离器是特定于任务的模块。
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.
