A Comparison of Human against Machine-Classification of Spatial Audio Scenes in Binaural Recordings of Music

Machine learning-based infant crying interpretation

Crying is an inevitable character trait that occurs throughout the growth of infants, under conditions where the caregiver may have difficulty interpreting the underlying cause of the cry. Crying can be treated as an audio signal that carries a message about the infant's state, such as discomfort, hunger, and sickness. The primary infant caregiver requires traditional ways of understanding these feelings. Failing to understand them correctly can cause severe problems. Several methods attempt to solve this problem; however, proper audio feature representation and classifiers are necessary for better results. This study uses time-, frequency-, and time-frequency-domain feature representations to gain in-depth information from the data. The time-domain features include zero-crossing rate (ZCR) and root mean square (RMS), the frequency-domain feature includes the Mel-spectrogram, and the time-frequency-domain feature includes Mel-frequency cepstral coefficients (MFCCs). Moreover, time-series imaging algorithms are applied to transform 20 MFCC features into images using different algorithms: Gramian angular difference fields, Gramian angular summation fields, Markov transition fields, recurrence plots, and RGB GAF. Then, these features are provided to different machine learning classifiers, such as decision tree, random forest, K nearest neighbors, and bagging. The use of MFCCs, ZCR, and RMS as features achieved high performance, outperforming state of the art (SOTA). Optimal parameters are found via the grid search method using 10-fold cross-validation. Our MFCC-based random forest (RF) classifier approach achieved an accuracy of 96.39%, outperforming SOTA, the scalogram-based shuffleNet classifier, which had an accuracy of 95.17%.

Comparative Analysis of Human Operators and Advanced Technologies in the Visual Inspection of Aero Engine Blades

Background—Aircraft inspection is crucial for safe flight operations and is predominantly performed by human operators, who are unreliable, inconsistent, subjective, and prone to err. Thus, advanced technologies offer the potential to overcome those limitations and improve inspection quality. Method—This paper compares the performance of human operators with image processing, artificial intelligence software and 3D scanning for different types of inspection. The results were statistically analysed in terms of inspection accuracy, consistency and time. Additionally, other factors relevant to operations were assessed using a SWOT and weighted factor analysis. Results—The results show that operators’ performance in screen-based inspection tasks was superior to inspection software due to their strong cognitive abilities, decision-making capabilities, versatility and adaptability to changing conditions. In part-based inspection however, 3D scanning outperformed the operator while being significantly slower. Overall, the strength of technological systems lies in their consistency, availability and unbiasedness. Conclusions—The performance of inspection software should improve to be reliably used in blade inspection. While 3D scanning showed the best results, it is not always technically feasible (e.g., in a borescope inspection) nor economically viable. This work provides a list of evaluation criteria beyond solely inspection performance that could be considered when comparing different inspection systems.

Spatial Audio Scene Characterization (SASC): Automatic Localization of Front-, Back-, Up-, and Down-Positioned Music Ensembles in Binaural Recordings

The automatic localization of audio sources distributed symmetrically with respect to coronal or transverse planes using binaural signals still poses a challenging task, due to the front–back and up–down confusion effects. This paper demonstrates that the convolutional neural network (CNN) can be used to automatically localize music ensembles panned to the front, back, up, or down positions. The network was developed using the repository of the binaural excerpts obtained by the convolution of multi-track music recordings with the selected sets of head-related transfer functions (HRTFs). They were generated in such a way that a music ensemble (of circular shape in terms of its boundaries) was positioned in one of the following four locations with respect to the listener: front, back, up, and down. According to the obtained results, CNN identified the location of the ensembles with the average accuracy levels of 90.7% and 71.4% when tested under the HRTF-dependent and HRTF-independent conditions, respectively. For HRTF-dependent tests, the accuracy decreased monotonically with the increase in the ensemble size. A modified image occlusion sensitivity technique revealed selected frequency bands as being particularly important in terms of the localization process. These frequency bands are largely in accordance with the psychoacoustical literature.

Special Issue on Deep Learning for Applications in Acoustics: Modeling, Synthesis, and Listening

The recent introduction of Deep Learning has led to a vast array of breakthroughs in many fields of science and engineering [...]