Gamma Boltzmann Machine for Audio Modeling
Gamma Boltzmann Machine for Audio Modeling
复制标题
用于音频建模的伽马玻尔兹曼机
DOI：
10.1109/taslp.2021.3095656
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Yatabe Kohei
中科院分区：
文献类型：
--
作者：
Nakashika Toru;Yatabe Kohei
本文提出了一种基于能量的概率模型，考虑线性和对数尺度处理非负数据。在音频应用中，包括频谱图在内的时频表示的幅度被认为是最重要的特征之一。这种基于幅度的特征已被广泛用于基于学习的音频处理。由于对数尺度在听觉感知方面很重要，因此通常使用对数函数计算特征。也就是说，在特征的计算中应用对数函数，使得学习机不必显式地对对数尺度进行建模。我们认为在一个不同的方式，并提出了一个限制玻尔兹曼机（RBM），同时模型的线性和对数幅度谱。RBM是一种随机神经网络，可以在没有监督的情况下发现数据表示。为了管理线性和对数尺度，我们定义了一个基于这两种尺度的能量函数。该能量函数导致条件分布（给定隐藏单元的可观测数据的条件分布），其被写为伽马分布，因此所提出的RBM被称为伽马-伯努利RBM。通过语音表征实验，将该算法与普通的Gaussian-Bernoulli RBM进行了比较。客观评价和主观评价都说明了该模型的优点。
This paper presents an energy-based probabilistic model that handles nonnegative data in consideration of both linear and logarithmic scales. In audio applications, magnitude of time-frequency representation, including spectrogram, is regarded as one of the most important features. Such magnitude-based features have been extensively utilized in learning-based audio processing. Since a logarithmic scale is important in terms of auditory perception, the features are usually computed with a logarithmic function. That is, a logarithmic function is applied within the computation of features so that a learning machine does not have to explicitly model the logarithmic scale. We think in a different way and propose a restricted Boltzmann machine (RBM) that simultaneously models linear- and log-magnitude spectra. RBM is a stochastic neural network that can discover data representations without supervision. To manage both linear and logarithmic scales, we define an energy function based on both scales. This energy function results in a conditional distribution (of the observable data, given hidden units) that is written as the gamma distribution, and hence the proposed RBM is termed gamma-Bernoulli RBM. The proposed gamma-Bernoulli RBM was compared to the ordinary Gaussian-Bernoulli RBM by speech representation experiments. Both objective and subjective evaluations illustrated the advantage of the proposed model.