Exploiting Parallel Audio Recordings to Enforce Device Invariance in
                        CNN-based Acoustic Scene Classification

Primus, Paul; Eghbal-zadeh, Hamid; Eitelsebner, David; Koutini, Khaled; Arzt, Andreas; Widmer, Gerhard

doi:10.33682/v9qj-8954

Cited by 19 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our motivation is paying more attention to the hard-to-adapt domain. The improved loss function is defined as formula (10) shows. Where ui denotes the domain index of i-th sample, N denotes the number of data.…”

Section: Multi-classification Taskmentioning

confidence: 99%

“…Figure 1 shows that different recording devices lead to the change of data distribution. To address the problem of mismatched recording devices, many methods have been proposed, such as data augmentation [6,7], spectrum correction [8,9] and domain adaptation (DA) [10]. Although these methods got good performance, they trained with both labeled source-and target-domain samples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification

Yang¹,

Wang²,

Zou³

2021

Interspeech 2021

View full text Add to dashboard Cite

It is well known that the mismatch between training (source) and test (target) data distribution will significantly decrease the performance of acoustic scene classification (ASC) systems. To address this issue, domain adaptation (DA) is one solution and many unsupervised DA methods have been proposed. These methods focus on a scenario of single source domain to single target domain. However, we will face such problem that test data comes from multiple target domains. This problem can be addressed by producing one model per target domain, but this solution is too costly. In this paper, we propose a novel unsupervised multi-target domain adaption (MTDA) method for ASC, which can adapt to multiple target domains simultaneously and make use of the underlying relation among multiple domains. Specifically, our approach combines traditional adversarial adaptation with two novel discriminator tasks that learns a common subspace shared by all domains. Furthermore, we propose to divide the target domain into the easy-to-adapt and hard-to-adapt domain, which enables the system to pay more attention to hard-to-adapt domain in training. The experimental results on the DCASE 2020 Task 1-A dataset and the DCASE 2019 Task 1-B dataset show that our proposed method significantly outperforms the previous unsupervised DA methods.

show abstract

Section: Multi-classification Taskmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification

Yang¹,

Wang²,

Zou³

2021

Interspeech 2021

View full text Add to dashboard Cite

show abstract

“…One approach to achieve such an alignment is to standardize data per frequency band independently per domain [11] or by matching the band-wise statistics between domains [12]. In general, such an adaptation can not only be performed on the feature-level but also using internal hidden layer activations [13].…”

Section: Related Workmentioning

confidence: 99%

Towards Audio Domain Adaptation for Acoustic Scene Classification using Disentanglement Learning

Abeßer¹,

Müller²

2021

Preprint

View full text Add to dashboard Cite

The deployment of machine listening algorithms in real-life applications is often impeded by a domain shift caused for instance by different microphone characteristics. In this paper, we propose a novel domain adaptation strategy based on disentanglement learning. The goal is to disentangle task-specific and domain-specific characteristics in the analyzed audio recordings. In particular, we combine two strategies: First, we apply different binary masks to internal embedding representations and, second, we suggest a novel combination of categorical cross-entropy and variance-based losses. Our results confirm the disentanglement of both tasks on an embedding level but show only minor improvement in the acoustic scene classification performance, when training data from both domains can be used. As a second finding, we can confirm the effectiveness of a state-of-the-art unsupervised domain adaptation strategy, which performs across-domain adaptation on a feature-level instead.

show abstract

“…For example, we can use multiple features obtained from the source domain to extract target domain features by establishing saddle points to improve the classification accuracy [20]. The last one is model adaptation, which modifies the loss function in the source domain to match the loss in the target domain [21].…”

Section: Introductionmentioning

confidence: 99%

Feature Alignment for Robust Acoustic Scene Classification Across Devices

Zhao

Kong

Song

et al. 2022

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

This letter presents a feature alignment method for domain adaptive Acoustic Scene Classification (ASC) across recording devices. First, we design a two-stream network, in which each stream processes two features, i.e., Log-Mel spectrogram and delta-deltas, using two sub-networks. Second, we investigate different loss functions for feature alignment between the feature maps obtained by the source and target domains. Last, we present an alternate training strategy to deal with the data imbalance problem between paired and unpaired samples. The experimental results obtained on the DCASE benchmarks demonstrate the effectiveness and superiority of the proposed method. The source code of the proposed method is available at https://github.com/Jingqiao-Zhao/FAASC.

show abstract

Exploiting Parallel Audio Recordings to Enforce Device Invariance in CNN-based Acoustic Scene Classification

Cited by 19 publications

References 0 publications

Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification

Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification

Towards Audio Domain Adaptation for Acoustic Scene Classification using Disentanglement Learning

Feature Alignment for Robust Acoustic Scene Classification Across Devices

Contact Info

Product

Resources

About