Dimensional feature weighting utilizing multiple kernel learning for single-channel talker location discrimination using the acoustic transfer function

Takashima, Ryoichi; Takiguchi, Tetsuya; Ariki, Yasuo

doi:10.1121/1.4773255

Cited by 4 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…τ is the index number of frequency bins of the short-term linear spectra in the n-th frame sound signal sequence, and L ′ express the length of the ATF in the STFT domain. However, the cost of such solution to estimate the frame sequence of the ATF is quite expensive [41]. Therefore, the estimated components of the ATF are too complex and it is difficult to deal with those parameters for this task.…”

Section: Estimation Of Source Distance With the Proposed Acoustic Modelmentioning

confidence: 99%

Multimodal fusion for indoor sound source localization

Chen

Takashima

Guo

et al. 2021

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Section: Estimation Of Source Distance With the Proposed Acoustic Modelmentioning

confidence: 99%

Multimodal fusion for indoor sound source localization

Chen

Takashima

Guo

et al. 2021

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

“…From the artificial pinna, the sound source elevation is estimated by using the propagation transfer function and the neural network classifier [ 23 ]. The characteristics of indoor speech propagation were utilized for non-structural ML within a limited situation [ 24 , 25 , 26 ]. The parabolic structure with cepstral speech parameters was explored for position-dependent indoor ML [ 27 ].…”

Section: Introductionmentioning

confidence: 99%

Monaural Sound Localization Based on Reflective Structure and Homomorphic Deconvolution

Park

Choi

Kim

2017

Sensors

View full text Add to dashboard Cite

The asymmetric structure around the receiver provides a particular time delay for the specific incoming propagation. This paper designs a monaural sound localization system based on the reflective structure around the microphone. The reflective plates are placed to present the direction-wise time delay, which is naturally processed by convolutional operation with a sound source. The received signal is separated for estimating the dominant time delay by using homomorphic deconvolution, which utilizes the real cepstrum and inverse cepstrum sequentially to derive the propagation response’s autocorrelation. Once the localization system accurately estimates the information, the time delay model computes the corresponding reflection for localization. Because of the structure limitation, two stages of the localization process perform the estimation procedure as range and angle. The software toolchain from propagation physics and algorithm simulation realizes the optimal 3D-printed structure. The acoustic experiments in the anechoic chamber denote that 79.0% of the study range data from the isotropic signal is properly detected by the response value, and 87.5% of the specific direction data from the study range signal is properly estimated by the response time. The product of both rates shows the overall hit rate to be 69.1%.

show abstract

“…The directivity pattern of the head-related transfer function was improved by exploring the various structures around the microphone [ 24 ]. Non-structural ML actively utilized the characteristics of indoor speech propagation for situation-related localization [ 25 , 26 , 27 ]. The indoor condition exhibited position-dependent Cepstral and speech parameters, which could be further enhanced by the parabolic reflection structure for the ML system [ 28 ].…”

Section: Introductionmentioning

confidence: 99%

Near-Field Sound Localization Based on the Small Profile Monaural Structure

Kim

2015

Sensors

View full text Add to dashboard Cite

The acoustic wave around a sound source in the near-field area presents unconventional properties in the temporal, spectral, and spatial domains due to the propagation mechanism. This paper investigates a near-field sound localizer in a small profile structure with a single microphone. The asymmetric structure around the microphone provides a distinctive spectral variation that can be recognized by the dedicated algorithm for directional localization. The physical structure consists of ten pipes of different lengths in a vertical fashion and rectangular wings positioned between the pipes in radial directions. The sound from an individual direction travels through the nearest open pipe, which generates the particular fundamental frequency according to the acoustic resonance. The Cepstral parameter is modified to evaluate the fundamental frequency. Once the system estimates the fundamental frequency of the received signal, the length of arrival and angle of arrival (AoA) are derived by the designed model. From an azimuthal distance of 3–15 cm from the outer body of the pipes, the extensive acoustic experiments with a 3D-printed structure show that the direct and side directions deliver average hit rates of 89% and 73%, respectively. The closer positions to the system demonstrate higher accuracy, and the overall hit rate performance is 78% up to 15 cm away from the structure body.

show abstract

Dimensional feature weighting utilizing multiple kernel learning for single-channel talker location discrimination using the acoustic transfer function

Cited by 4 publications

References 13 publications

Multimodal fusion for indoor sound source localization

Multimodal fusion for indoor sound source localization

Monaural Sound Localization Based on Reflective Structure and Homomorphic Deconvolution

Near-Field Sound Localization Based on the Small Profile Monaural Structure

Contact Info

Product

Resources

About