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 1906.08043eess.AScs.CLcs.SD
收藏

用于语音识别的实数到H空间编码器

Real to H-space Encoder for Speech Recognition

Avignon Université(阿维尼翁大学);ORKIS(ORKIS公司);McGill University(麦吉尔大学)

Titouan Parcollet, Mohamed Morchid, Georges Linarès, Renato De Mori

摘要:

深度神经网络(DNN)和更准确地说,递归神经网络(RNN)是现代自动语音识别系统的核心,因为它们处理输入序列的效率。最近,已经表明,基于多维代数的不同输入表示,例如复数和四元数,能够通过优于常见的实值NN,为神经网络带来更自然,压缩和强大的输入信号表示。事实上,四元数值神经网络(QNN)可以更好地学习内部依赖关系,例如特定时间帧的Mel滤波器组值与其时间导数之间的关系,以及全局依赖关系,描述时间帧之间存在的关系。尽管如此,QNN仅限于四元数值输入信号,并且很难从这种具有实值输入数据的强大表示中受益。本文提出通过引入实数到四元数编码器来解决这个弱点,该编码器允许QNN处理任何一维输入特征,例如用于自动语音识别的传统Mel滤波器组。

英文摘要:

Deep neural networks (DNNs) and more precisely recurrent neural networks (RNNs) are at the core of modern automatic speech recognition systems, due to their efficiency to process input sequences. Recently, it has been shown that different input representations, based on multidimensional algebras, such as complex and quaternion numbers, are able to bring to neural networks a more natural, compressive and powerful representation of the input signal by outperforming common real-valued NNs. Indeed, quaternion-valued neural networks (QNNs) better learn both internal dependencies, such as the relation between the Mel-filter-bank value of a specific time frame and its time derivatives, and global dependencies, describing the relations that exist between time frames. Nonetheless, QNNs are limited to quaternion-valued input signals, and it is difficult to benefit from this powerful representation with real-valued input data. This paper proposes to tackle this weakness by introducing a real-to-quaternion encoder that allows QNNs to process any one dimensional input features, such as traditional Mel-filter-banks for automatic 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。

↑