Representation of complex spectrogram via phase conversion

Yatabe, Kohei; Masuyama, Yoshiaki; Kusano, Tsubasa; Oikawa, Yasuhiro

doi:10.1250/ast.40.170

Cited by 30 publications

(26 citation statements)

References 47 publications

(67 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. To avoid the sensitivity problem, DNNs estimate phase derivatives instead of phase itself since phase derivatives have a relation to the amplitude spectrogram [24][25][26]. Then, the proposed method recursively reconstructs phase from the estimated phase derivatives, which is named RPU.…”

Section: Amplitude Spectrogrammentioning

confidence: 99%

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Masuyama

Yatabe

Oikawa

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Phase reconstruction, which estimates phase from a given amplitude spectrogram, is an active research field in acoustical signal processing with many applications including audio synthesis. To take advantage of rich knowledge from data, several studies presented deep neural network (DNN)-based phase reconstruction methods. However, the training of a DNN for phase reconstruction is not an easy task because phase is sensitive to the shift of a waveform. To overcome this problem, we propose a DNN-based two-stage phase reconstruction method. In the proposed method, DNNs estimate phase derivatives instead of phase itself, which allows us to avoid the sensitivity problem. Then, phase is recursively estimated based on the estimated derivatives, which is named recurrent phase unwrapping (RPU). The experimental results confirm that the proposed method outperformed the direct phase estimation by a DNN.

show abstract

Section: Amplitude Spectrogrammentioning

confidence: 99%

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Masuyama

Yatabe

Oikawa

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, the phase-conversion method that modifies the time sensitivity of phases was proposed [1,2]. In the present study, we applied it to speech analyses and verified its efficiency.…”

Section: Introductionmentioning

confidence: 74%

Visualization of phase characteristics of a vocal tract using phase-conversion method with electroglottograph signals

Okada

2021

Acoust. Sci. & Tech.

View full text Add to dashboard Cite

“…As the conventional method, DNN-based speech enhancement in STFT domain was considered. STFT with the 512 points (32 ms) Hann window, 256 points time-shifting and 512 points discrete Fourier transform length was used, and the inverse STFT was implemented by its canonical dual [29] to make the STFT as perfect reconstruction filterbank. To estimate real-valued T-F mask, U-Net illustrated in Fig.…”

Section: Dnn Architecture Loss Function and Training Setupmentioning

confidence: 99%

Invertible DNN-Based Nonlinear Time-Frequency Transform for Speech Enhancement

Takeuchi

Yatabe

Oikawa

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We propose an end-to-end speech enhancement method with trainable time-frequency (T-F) transform based on invertible deep neural network (DNN). The resent development of speech enhancement is brought by using DNN. The ordinary DNN-based speech enhancement employs T-F transform, typically the short-time Fourier transform (STFT), and estimates a T-F mask using DNN. On the other hand, some methods have considered end-to-end networks which directly estimate the enhanced signals without T-F transform. While end-to-end methods have shown promising results, they are black boxes and hard to understand. Therefore, some end-to-end methods used a DNN to learn the linear T-F transform which is much easier to understand. However, the learned transform may not have a

show abstract

Representation of complex spectrogram via phase conversion

Cited by 30 publications

References 47 publications

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Visualization of phase characteristics of a vocal tract using phase-conversion method with electroglottograph signals

Invertible DNN-Based Nonlinear Time-Frequency Transform for Speech Enhancement

Contact Info

Product

Resources

About