| Paper ID | F-1-3.5 |
| Paper Title |
Gamma Boltzmann Machine for Simultaneously Modeling Linear- and Log-amplitude Spectra |
| Authors |
Toru Nakashika, The University of Electro-Communications, Japan; Kohei Yatabe, Waseda University, Japan |
| Session |
F-1-3: Speech Enhancement 1 |
| Time | Tuesday, 08 December, 17:15 - 19:15 |
| Presentation Time: | Tuesday, 08 December, 18:15 - 18:30 Check your Time Zone |
|
All times are in New Zealand Time (UTC +13) |
| Topic |
Speech, Language, and Audio (SLA): |
| Abstract |
In audio applications, one of the most important rep- resentations of audio signals is the amplitude spectrogram. It is utilized in many machine-learning-based information processing methods including the ones using the restricted Boltzmann ma- chines (RBM). However, the ordinary Gaussian-Bernoulli RBM (the most popular RBM among its variations) cannot directly handle amplitude spectra because the Gaussian distribution is a symmetric model allowing negative values which never appear in the amplitude. In this paper, after proposing a general gamma Boltzmann machine, we propose a practical model called the gamma-Bernoulli RBM that simultaneously handles both linear- and log-amplitude spectrograms. Its conditional distribution of the observable data is given by the gamma distribution, and thus the proposed RBM can naturally handle the data represented by positive numbers as the amplitude spectra. It can also treat amplitude in the logarithmic scale which is important for audio signals from the perceptual point of view. The advantage of the proposed model compared to the ordinary Gaussian-Bernoulli RBM was confirmed by PESQ and MSE in the experiment of representing the amplitude spectrograms of speech signals. |