This paper investigates a self-adaptation method for speech enhancement using auxiliary speaker-aware features; we extract a speaker representation used for adaptation directly from the test utterance. Conventional studies of deep neural network (DNN)-based speech enhancement mainly focus on building a speaker independent model. Meanwhile, in speech applications including speech recognition and synthesis, it is known that model adaptation to the target speaker improves the accuracy. Our research question is whether a DNN for speech enhancement can be adopted to unknown speakers without any auxiliary guidance signal in test-phase. To achieve this, we adopt multi-task learning of speech enhancement and speaker identification, and use the output of the final hidden layer of speaker identification branch as an auxiliary feature. In addition, we use multi-head self-attention for capturing long-term dependencies in the speech and noise. Experimental results on a public dataset show that our strategy achieves the state-of-the-art performance and also outperform conventional methods in terms of subjective quality.

This paper presents a novel phase reconstruction method (only from a given amplitude spectrogram) by combining a signal-processingbased approach and a deep neural network (DNN). To retrieve a time-domain signal from its amplitude spectrogram, the corresponding phase is required. One of the popular phase reconstruction methods is the Griffin-Lim algorithm (GLA), which is based on the redundancy of the short-time Fourier transform. However, GLA often involves many iterations and produces low-quality signals owing to the lack of prior knowledge of the target signal. In order to address these issues, in this study, we propose an architecture which stacks a sub-block including two GLA-inspired fixed layers and a DNN. The number of stacked sub-blocks is adjustable, and we can trade the performance and computational load based on requirements of applications. The effectiveness of the proposed method is investigated by reconstructing phases from amplitude spectrograms of speeches.

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.

Recovering a signal from its amplitude spectrogram, or phase recovery, exhibits many applications in acoustic signal processing. When only an amplitude spectrogram is available and no explicit information is given for the phases, the Griffin-Lim algorithm (GLA) is one of the most utilized methods for phase recovery. However, GLA often requires many iterations and results in low perceptual quality in some cases. In this letter, we propose two novel algorithms based on GLA and the alternating direction method of multipliers (ADMM) for better recovery with fewer iteration. Some interpretation of the existing methods and their relation to the proposed method are also provided. Evaluations are performed with both objective measure and subjective test.

Sound-field imaging, the visualization of spatial and temporal distribution of acoustical properties such as sound pressure, is useful for understanding acoustical phenomena. This study investigated the use of parallel phase-shifting interferometry (PPSI) with a high-speed polarization camera for imaging a sound field, particularly high-speed imaging of propagating sound waves. The experimental results showed that the instantaneous sound field, which was generated by ultrasonic transducers driven by a pure tone of 40 kHz, was quantitatively imaged. Hence, PPSI can be used in acoustical applications requiring spatial information of sound pressure.

No abstract

As importance of the phase of complex spectrogram has been recognized widely, many techniques have been proposed for handling it. However, several definitions and terminologies for the same concept can be found in the literature, which has confused beginners. In this paper, two major definitions of the short-time Fourier transform and their phase conventions are summarized to alleviate such complication. A phase-aware signal-processing scheme based on phase conversion is also introduced with a set of executable MATLAB functions (https://doi.org/10/c3qb).

Multichannel audio blind source separation (BSS) in the determined situation (the number of microphones is equal to that of the sources), or determined BSS, is performed by multichannel linear filtering in the time-frequency domain to handle the convolutive mixing process. Ordinarily, the filter treats each frequency independently, which causes the wellknown permutation problem, i.e., the problem of how to align the frequency-wise filters so that each separated component is correctly assigned to the corresponding sources. In this paper, it is shown that the general property of the time-frequency-domain representation called spectrogram consistency can be an assistant for solving the permutation problem.