arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2005.08566eess.AScs.LGcs.SDstat.ML
收藏

基于四元数神经网络的多通道远程语音识别

Quaternion Neural Networks for Multi-channel Distant Speech Recognition

University of Oxford(牛津大学);Samsung AI(三星人工智能);Avignon University(阿维尼翁大学)

Xinchi Qiu, Titouan Parcollet, Mirco Ravanelli, Nicholas Lane, Mohamed Morchid

摘要:

尽管在自动语音识别(ASR)方面取得了重大进展,但由于噪声和混响,远距离ASR仍然具有挑战性。缓解此问题的常见方法包括为记录设备配备多个麦克风,这些麦克风从不同角度捕获声学场景。这些多通道音频记录包含每个信号之间的特定内部关系。在本文中,我们提出捕捉这些内部和内部结构的依赖关系与四元数神经网络,它可以联合处理多个信号作为整个四元数实体。四元数代数将标准的点积替换为Hamilton点积,从而提供了一种简单而优雅的方式来建模元素之间的依赖关系。然后,四元数层与递归神经网络耦合,该网络可以在时域中学习长期依赖关系。我们证明了在级联多通道语音信号上训练的四元数长短期记忆神经网络(QLSTM)在多通道远程语音识别的两个不同任务上优于等效的实值LSTM。

英文摘要:

Despite the significant progress in automatic speech recognition (ASR), distant ASR remains challenging due to noise and reverberation. A common approach to mitigate this issue consists of equipping the recording devices with multiple microphones that capture the acoustic scene from different perspectives. These multi-channel audio recordings contain specific internal relations between each signal. In this paper, we propose to capture these inter- and intra- structural dependencies with quaternion neural networks, which can jointly process multiple signals as whole quaternion entities. The quaternion algebra replaces the standard dot product with the Hamilton one, thus offering a simple and elegant way to model dependencies between elements. The quaternion layers are then coupled with a recurrent neural network, which can learn long-term dependencies in the time domain. We show that a quaternion long-short term memory neural network (QLSTM), trained on the concatenated multi-channel speech signals, outperforms equivalent real-valued LSTM on two different tasks of multi-channel distant speech recognition.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑