Dance Tempo Estimation Using a Single Leg-Attached 3D Accelerometer

We present a methodology that enables dance tempo estimation through the acquisition of 3D accelerometer signals using a single wearable inertial device positioned on the dancer’s leg. Our tempo estimation method is based on enhanced multiple resonators, implemented with comb feedback filters. To validate the methodology, we focus on the versatile solo jazz dance style. Including a variety of dance moves, with different leg activation patterns and rhythmical variations, solo jazz provides for a highly critical validation environment. We consider 15 different solo jazz dance moves, with different leg activation patterns, assembled in a sequence of 5 repetitions of each, giving 65 moves altogether. A professional and a recreational dancer performed this assembly in a controlled environment, following eight dancing tempos, dictated by a metronome, and ranging from 80 bpm to 220 bpm with 20 bpm increment steps. We show that with appropriate enhancements and using single leg signals, the comb filter bank provides for accurate dance tempo estimates for all moves and rhythmical variations considered. Dance tempo estimates for the overall assembles match strongly the dictated tempo—the difference being at most 1 bpm for all measurement instances is within the limits of the established beat onset stability of the used metronome. Results further show that this accuracy is achievable for shorter dancing excerpts, comprising four dance moves, corresponding to one music phrase, and as such enables real-time feedback. By providing for a dancer’s tempo quality and consistency assessment, the presented methodology has the potential of supporting the learning process, classifying individual level of experience, and assessing overall performance. It is extendable to other dance styles and sport motion in general where cyclical patterns occur.

Volitional EMG Estimation Method during Functional Electrical Stimulation by Dual-Channel Surface EMGs

We propose a novel dual-channel electromyography (EMG) spatio-temporal differential (DESTD) method that can estimate volitional electromyography (vEMG) signals during time-varying functional electrical stimulation (FES). The proposed method uses two pairs of EMG signals from the same stimulated muscle to calculate the spatio-temporal difference between the signals. We performed an experimental study with five healthy participants to evaluate the vEMG signal estimation performance of the DESTD method and compare it with that of the conventional comb filter and Gram–Schmidt methods. The normalized root mean square error (NRMSE) values between the semi-simulated raw vEMG signal and vEMG signals which were estimated using the DESTD method and conventional methods, and the two-tailed t-test and analysis of variance were conducted. The results showed that under the stimulation of the gastrocnemius muscle with rapid and dynamically modulated stimulation intensity, the DESTD method had a lower NRMSE compared to the conventional methods (p< 0.01) for all stimulation intensities (maximum 5, 10, 15, and 20 mA). We demonstrated that the DESTD method could be applied to wearable EMG-controlled FES systems because it estimated vEMG signals more effectively compared to the conventional methods under dynamic FES conditions and removed unnecessary FES-induced EMG signals.

Improving Comb Aliasing Rejection Using Sharpening of Modified Comb

This paper presents an efficient method to improve the comb aliasing rejection in a comb decimation filter without increasing its passband droop. This problem is important since aliasing and comb passband droop may deteriorate the decimated signal. We propose here to apply sharpening of the modified comb in the second stage of a two-stage comb structure. The modified comb is obtained by decreasing the middle coefficient of the impulse response of the cascade of two combs by 1/2. The sharpening polynomial with the first order tangencies is used here. As a result, the comb folding bands, where the aliasing occur, become wider and with an increased attenuation in comparison with the original comb filter. However, this improvement in the folding bands did not result in an increased passband droop. The compensator from literature is used to further decrease passband droop. The method is illustrated with examples and compared with the original comb and the methods proposed in literature for increasing aliasing rejection.

Neural Comb Filtering using Sliding Window Attention Network for Speech Enhancement

<div>In this paper, we demonstrate the significance of restoring harmonics of the fundamental frequency (pitch) in deep neural network (DNN) based speech enhancement. We propose a sliding-window attention network to regress the spectral magnitude mask (SMM) from the noisy speech signal. Even though the network parameters can be estimated by minimizing the mask loss, it does not restore the pitch harmonics, especially at higher frequencies. In this paper, we propose to restore the pitch harmonics in the spectral domain by minimizing cepstral loss around the pitch peak. The network parameters are estimated using a combination of the mask loss and cepstral loss. The proposed network architecture functions like an adaptive comb filter on voiced segments, and emphasizes the pitch harmonics in the speech spectrum. The proposed approach achieves comparable performance with the state-of-the-art methods with much lesser computational complexity.</div>

Neural Comb Filtering using Sliding Window Attention Network for Speech Enhancement

<div>In this paper, we demonstrate the significance of restoring harmonics of the fundamental frequency (pitch) in deep neural network (DNN) based speech enhancement. We propose a sliding-window attention network to regress the spectral magnitude mask (SMM) from the noisy speech signal. Even though the network parameters can be estimated by minimizing the mask loss, it does not restore the pitch harmonics, especially at higher frequencies. In this paper, we propose to restore the pitch harmonics in the spectral domain by minimizing cepstral loss around the pitch peak. The network parameters are estimated using a combination of the mask loss and cepstral loss. The proposed network architecture functions like an adaptive comb filter on voiced segments, and emphasizes the pitch harmonics in the speech spectrum. The proposed approach achieves comparable performance with the state-of-the-art methods with much lesser computational complexity.</div>