Phase Reconstruction from Amplitude Spectrograms Based on Von-Mises-Distribution Deep Neural Network

By Shinnosuke Takamichi, Yuki Saito, Norihiro Takamune, Daichi Kitamura, Hiroshi Saruwatari. 2018

47 citations recorded by the source corpus.

This paper presents a deep neural network (DNN)-based phase reconstruction from amplitude spectrograms. In audio signal and speech processing, the amplitude spectrogram is often used for processing, and the corresponding phase spectrogram is reconstructed from the amplitude spectrogram on the basis of the Griffin-Lim method. However, the Griffin-Lim method causes unnatural artifacts in synthetic speech. Addressing this problem, we introduce the von-Mises-distribution DNN for phase reconstruction. The DNN is a generative model having the von Mises distribution that can model distributions of a periodic variable such as a phase, and the model parameters of the DNN are estimated on the basis of the maximum likelihood criterion. Furthermore, we propose a group-delay loss for DNN training to make the predicted group delay close to a natural group delay. The experimental results demonstrate that 1) the trained DNN can predict group delay accurately more than phases themselves, and 2) our phase reconstruction methods achieve better speech quality than the conventional Griffin-Lim method.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Ke Tan
2. Deliang Wang
3. Zhong-Qiu Wang
4. Hubert Siuzdak
5. Arun Narayanan
6. Yuxuan Wang
7. Bo Li
8. Yossi Adi

In Signal Processing

1. Ke Tan
2. Zhong-Qiu Wang
3. Deliang Wang
4. Hubert Siuzdak
5. Yuxuan Wang
6. Jonathan Le Roux
7. Arun Narayanan
8. Yossi Adi

Citations in the corpus

• Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (2023, methodology): “Several methods have been proposed for reconstructing phase using deep neural networks, including likelihood-based approaches (Takamichi et al., 2018) and GANs (Oyamada et al.”
• Deep Learning Based Phase Reconstruction for Speaker Separation: A Trigonometric Perspective (2018, result): “This distinguishes our approach from studies that directly predict unbounded or unconstrained phase differences [22], [23], clean phases [24] and real and imaginary components of target sources [25], [26], or fully complex neural network approaches [27].”