Note on Binaural Masking‐Level Differences as a Function of the Interaural Correlation of the Masking Noise

1964 ◽  
Vol 36 (9) ◽  
pp. 1613-1617 ◽  
Author(s):  
N. I. Durlach
Author(s):  
Claire Guang ◽  
Emmett Lefkowitz ◽  
Naseem Dillman-Hasso ◽  
Violet Brown ◽  
Julia Strand
Keyword(s):  

1962 ◽  
Vol 34 (10) ◽  
pp. 1658-1659 ◽  
Author(s):  
Lloyd A. Jeffress ◽  
Donald E. Robinson

2021 ◽  
Author(s):  
Jacob Hendriks ◽  
Patrick Dumond

Abstract This paper demonstrates various data augmentation techniques that can be used when working with limited run-to-failure data to estimate health indicators related to the remaining useful life of roller bearings. The PRONOSTIA bearing prognosis dataset is used for benchmarking data augmentation techniques. The input to the networks are multi-dimensional frequency representations obtained by combining the spectra taken from two accelerometers. Data augmentation techniques are adapted from other machine learning fields and include adding Gaussian noise, region masking, masking noise, and pitch shifting. Augmented datasets are used in training a conventional CNN architecture comprising two convolutional and pooling layer sequences with batch normalization. Results from individually separating each bearing’s data for the purpose of validation shows that all methods, except pitch shifting, give improved validation accuracy on average. Masking noise and region masking both show the added benefit of dataset regularization by giving results that are more consistent after repeatedly training each configuration with new randomly generated augmented datasets. It is shown that gradually deteriorating bearings and bearings with abrupt failure are not treated significantly differently by the augmentation techniques.


1993 ◽  
Vol 36 (4) ◽  
pp. 799-807 ◽  
Author(s):  
Gerald A. Studebaker ◽  
Robert L. Sherbecoe ◽  
Christine Gilmore

Frequency-importance and transfer functions for the Auditec of St. Louis recordings of the NU-6 word test are reported. The functions were derived from the word recognition scores of 24 subjects with normal hearing who were tested under 128 conditions of filtering and talker-spectrum-matched noise. The importance function was broader and had a lower midpoint than the NU-6 importance function reported by Schum, Matthews, and Lee (1991), but still displayed a bimodal shape. The transfer function was steeper than the transfer function reported by Schum et al., but comparable in slope to the transfer function for low-context CNC words reported by Bell, Dirks, and Trine (1992). Results from a limited set of conditions presented in quiet suggest that the use of masking noise was partly responsible for the dissimilar importance and transfer functions obtained by Schum et al. and this study. Differences in the equipment used in each experiment and in the methods used to analyze the data appear to have contributed as well.


2016 ◽  
Vol 55 (8) ◽  
pp. 463-471 ◽  
Author(s):  
Harvey Dillon ◽  
Elizabeth Francis Beach ◽  
John Seymour ◽  
Lyndal Carter ◽  
Maryanne Golding

Sign in / Sign up

Export Citation Format

Share Document