Single Pass Spectrogram Inversion

Beauregard, Gerald T.; Harish, M.; Wyse, Lonce

doi:10.1109/icdsp.2015.7251907

Cited by 36 publications

(44 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While phase reconstruction with observed noisy phase has been applied to speech enhancement successfully [8][9][10], phase reconstruction solely from a given amplitude spectrogram is still a challenging problem. To address this problem, various approaches have been studied including the consistency-based approach [11][12][13] and model-based approach [14]. While the former approach is based on only the property of the short-time Fourier transform (STFT) [15], the latter one explicitly uses a model of the target signal.…”

Section: Introductionmentioning

confidence: 99%

“…While the former approach is based on only the property of the short-time Fourier transform (STFT) [15], the latter one explicitly uses a model of the target signal. By taking the property of the target signal into account, the model-based approach has achieved better performance than the consistency-based one in many applications [14,16,17].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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)

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: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

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)

View full text Add to dashboard Cite

show abstract

“…The model is trained for ∼600k iterations with a batch size of 16 distributed across 4 GPUs with synchronous updates. We compare our results to conventional implementations of GL [1] and SPSI [3] with and without extra GL iterations.…”

Section: A Experimental Setupmentioning

confidence: 99%

Fast Spectrogram Inversion Using Multi-Head Convolutional Neural Networks

Arık

Jun

Diamos

2019

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

We propose the multi-head convolutional neural network (MCNN) for waveform synthesis from spectrograms. Nonlinear interpolation in MCNN is employed with transposed convolution layers in parallel heads. MCNN enables significantly better utilization of modern multi-core processors than commonly-used iterative algorithms like Griffin-Lim, and yields very fast (more than 300x real-time) runtime. For training of MCNN, we use a large-scale speech recognition dataset and losses defined on waveforms that are related to perceptual audio quality. We demonstrate that MCNN constitutes a very promising approach for high-quality speech synthesis, without any iterative algorithms or autoregression in computations.

show abstract

“…Section IV). We therefore generalize the approach from [21] to the nonstationary case and calculate the frequencies using quadratic interpolation.…”

Section: B Modificationmentioning

confidence: 99%

“…are introduced in order to tackle the challenges arising from the application of non-uniform NSGFs. Hence, the proposed algorithm relies on techniques such as phase locking [7], transient detection [20], and quadratic interpolation [21] and integrates new methods for dealing with attack transients (cf. Section V-B), for determining the phase values from frequencies estimated by quadratic interpolation (cf.…”

Section: Contributions To State Of the Artmentioning

confidence: 99%

A Phase Vocoder Based on Nonstationary Gabor Frames

Ottosen

Dörfler

2017

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-We propose a new algorithm for time stretching music signals based on the theory of nonstationary Gabor frames (NSGFs). The algorithm extends the techniques of the classical phase vocoder (PV) by incorporating adaptive timefrequency (TF) representations and adaptive phase locking. The adaptive TF representations imply good time resolution for the onsets of attack transients and good frequency resolution for the sinusoidal components. We estimate the phase values only at peak channels and the remaining phases are then locked to the values of the peaks in an adaptive manner. During attack transients we keep the stretch factor equal to one and we propose a new strategy for determining which channels are relevant for reinitializing the corresponding phase values. In contrast to previously published algorithms we use a non-uniform NSGF to obtain a low redundancy of the corresponding TF representation. We show that with just three times as many TF coefficients as signal samples, artifacts such as phasiness and transient smearing can be greatly reduced compared to the classical PV. The proposed algorithm is tested on both synthetic and real world signals and compared with state of the art algorithms in a reproducible manner.

show abstract

Single Pass Spectrogram Inversion

Cited by 36 publications

References 5 publications

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Phase Reconstruction Based On Recurrent Phase Unwrapping With Deep Neural Networks

Fast Spectrogram Inversion Using Multi-Head Convolutional Neural Networks

A Phase Vocoder Based on Nonstationary Gabor Frames

Contact Info

Product

Resources

About