Deep Convolutional Neural Networks and Data Augmentation for Acoustic Event Recognition

Takahashi, Naoya; Gygli, Michael; Pfister, Beat; Gool, Luc Van

doi:10.21437/interspeech.2016-805

Cited by 140 publications

(99 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We augment the data using standard techniques [6]: random cut of the 8-second training sample, random amplification of each source from (0.75, 1.25), random selection of the left or the right channel and shuffling of the sources between different tracks in half of the training batch. We train using RAdam [7] with the Lookahead optimizer [8] for a max of 250 iterations.…”

Section: Methodsmentioning

confidence: 99%

Meta-Learning Extractors for Music Source Separation

Samuel

Ganeshan

Naradowsky

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We propose a hierarchical meta-learning-inspired model for music source separation (Meta-TasNet) in which a generator model is used to predict the weights of individual extractor models. This enables efficient parameter-sharing, while still allowing for instrument-specific parameterization. Meta-TasNet is shown to be more effective than the models trained independently or in a multi-task setting, and achieve performance comparable with state-of-the-art methods. In comparison to the latter, our extractors contain fewer parameters and have faster run-time performance. We discuss important architectural considerations, and explore the costs and benefits of this approach.

show abstract

Section: Methodsmentioning

confidence: 99%

Meta-Learning Extractors for Music Source Separation

Samuel

Ganeshan

Naradowsky

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…This framework maps audio bases, extracted by non-negative matrix factorization (NMF), to the detected visual objects. In recent year, audio event detection (AED) [8,29,36] has received attention in the research community. Most of the AED methods locate audio events and then classify each event.…”

Section: Related Workmentioning

confidence: 99%

Watch, Listen and Tell: Multi-Modal Weakly Supervised Dense Event Captioning

Rahman

Sigal

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Multi-modal learning, particularly among imaging and linguistic modalities, has made amazing strides in many high-level fundamental visual understanding problems, ranging from language grounding to dense event captioning. However, much of the research has been limited to approaches that either do not take audio corresponding to video into account at all, or those that model the audiovisual correlations in service of sound or sound source localization. In this paper, we present the evidence, that audio signals can carry surprising amount of information when it comes to high-level visual-lingual tasks. Specifically, we focus on the problem of weakly-supervised dense event captioning in videos and show that audio on its own can nearly rival performance of a state-of-the-art visual model and, combined with video, can improve on the state-of-theart performance. Extensive experiments on the ActivityNet Captions dataset show that our proposed multi-modal approach outperforms state-of-the-art unimodal methods, as well as validate specific feature representation and architecture design choices.

show abstract

“…Meanwhile, tagging the sound events consumes amounts of manpower. To address this problem, two kinds of approaches have been proposed: one efficiently making use of limited data [15][16][17][18][19] and the other based on additional data [3,20,21]. In the Figure 1: Architecture of the end-to-end audio classification network.…”

Section: Related Workmentioning

confidence: 99%

“…first type of method, Takahashi et al [15] mixed two different sounds belonging to one class to extend the training distribution. Mixup [17] interpolated the training data and trained the model to output the mixing ratio.…”

Section: Related Workmentioning

confidence: 99%

An End-to-End Audio Classification System Based on Raw Waveforms and Mix-Training Strategy

Chen¹,

Jing²,

Chen³

et al. 2019

Interspeech 2019

View full text Add to dashboard Cite

Audio classification can distinguish different kinds of sounds, which is helpful for intelligent applications in daily life. However, it remains a challenging task since the sound events in an audio clip is probably multiple, even overlapping. This paper introduces an end-to-end audio classification system based on raw waveforms and mix-training strategy. Compared to human-designed features which have been widely used in existing research, raw waveforms contain more complete information and are more appropriate for multi-label classification. Taking raw waveforms as input, our network consists of two variants of ResNet structure which can learn a discriminative representation. To explore the information in intermediate layers, a multi-level prediction with attention structure is applied in our model. Furthermore, we design a mix-training strategy to break the performance limitation caused by the amount of training data. Experiments show that the mean average precision of the proposed audio classification system on Audio Set dataset is 37.2%. Without using extra training data, our system exceeds the state-of-the-art multi-level attention model.

show abstract

Deep Convolutional Neural Networks and Data Augmentation for Acoustic Event Recognition

Cited by 140 publications

References 33 publications

Meta-Learning Extractors for Music Source Separation

Meta-Learning Extractors for Music Source Separation

Watch, Listen and Tell: Multi-Modal Weakly Supervised Dense Event Captioning

An End-to-End Audio Classification System Based on Raw Waveforms and Mix-Training Strategy

Contact Info

Product

Resources

About