A hybrid approach with multi-channel i-vectors and convolutional neural networks for acoustic scene classification

Abstract

Abstract-In Acoustic Scene Classification (ASC) two major approaches have been followed . While one utilizes engineered features such as mel-frequency-cepstral-coefficients (MFCCs), the other uses learned features that are the outcome of an optimization algorithm. I-vectors are the result of a modeling technique that usually takes engineered features as input. It has been shown that standard MFCCs extracted from monaural audio signals lead to i-vectors that exhibit poor performance, especially on indoor acoustic scenes. At the same time, Convolutional Neural Networks (CNNs) are well known for their ability to learn features by optimizing their filters. They have been applied on ASC and have shown promising results. In this paper, we first propose a novel multi-channel i-vector extraction scheme for ASC, improving their performance on indoor and outdoor scenes. Second, we propose a CNN architecture that achieves promising ASC results. Further, we show that i-vectors and CNNs capture complementary information from acoustic scenes. Finally, we propose a hybrid system for ASC using multi-channel i-vectors and CNNs by utilizing a score fusion technique. Using our method, we participated in the ASC task of the DCASE-2016 challenge. Our hybrid approach achieved 1 st rank among 49 submissions, substantially improving the previous state of the art.

If structure can exclaim: a novel robotic-assisted percussion method for spatial bolt-ball joint looseness detection

“…Similar investigations have also been reported in previous literature. [47][48][49][50][51][52] Even though these studies [53][54][55][56][57][58][59][60] demonstrate that CNNs have significant capacity to process the MFCC and capture features for classification, some demerits limit the performance. Notably, current CNN-based methods require large amounts of data to train the models extensively, and the models need to relearn their inherent parameters to incorporate classifications when encountering new input.…”

Environmental Acoustics Modelling Techniques for Forest Monitoring

“…Concerning the experimented methodologies, the deep learning methods are predominant, with Feed-Forward Neural Networks and Convolutional Neural Networks (CNN) in the leading position. Many strategies are currently operating CNNs in conjunction with a variety of features, among which log-scaled mel-spectrograms [26]- [28], CNN-LTE [25] or in hybrid approaches [29]. These implementations outperformed the other attempts to approach EASR and SER tasks.…”

Aroma Release of Olfactory Displays Based on Audio-Visual Content

“…A small number of systems have used spatial features extracted from binaural recordings. In order to obtain the advantages from feature engineering approaches (i-vector) and feature learning methods (CNN), the authors in [13] proposed a multichannel i-vector by computing MFCC for both channels in the audio sample. In addition, they built a CNN model similar to VGG-net (invented by the Visual Geometry Group) architecture that takes spectrogram features as the input.…”