Robust Speaker Recognition Using Speech Enhancement And Attention Model

Shi, Yanpei; Huang, Qiang; Hain, Thomas

doi:10.48550/arxiv.2001.05031

Cited by 8 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CBAM refines the 3-dimensional feature map by combining channel attention module and spatial attention module in CNN architecture [19]. Motivated by CBAM, several studies have been conducted in speech processing field [23,24]. However, to our best knowledge, there was no attempt to apply CBAM for VAD.…”

Section: Attention Modulementioning

confidence: 99%

Dual Attention in Time and Frequency Domain for Voice Activity Detection

Lee¹,

Jung²,

Kim³

2020

Interspeech 2020

View full text Add to dashboard Cite

Voice activity detection (VAD) is a challenging task in low signal-to-noise ratio (SNR) environment, especially in nonstationary noise. To deal with this issue, we propose a novel attention module that can be integrated in Long Short-Term Memory (LSTM). Our proposed attention module refines each LSTM layer's hidden states so as to make it possible to adaptively focus on in both time and frequency domain. Experiments are conducted on various noisy conditions of Aurora 4. Our proposed method obtains the 95.58 % area under the ROC curve (AUC), achieving 22 % relative improvement compared to baseline, with only 2.44 % increase in the number of parameters. Besides, we utilize a focal loss for alleviating the performance degradation caused by imbalance between speech and non-speech sections in training sets. The results show that the focal loss can improve the performance in various imbalance situations compared to the cross entropy loss, a commonly used loss function in VAD.

show abstract

Section: Attention Modulementioning

confidence: 99%

Dual Attention in Time and Frequency Domain for Voice Activity Detection

Lee¹,

Jung²,

Kim³

2020

Interspeech 2020

View full text Add to dashboard Cite

show abstract

“…To learn a noise-invariant speaker embedding, adversarial training [17,18] and variability-invariant loss [19] are investigated. Also, joint training of speech enhancement network and speaker embedding network can improve the ASV performance under noisy conditions [20,21,22]. For deep speaker modeling with microphone array, a multichannel training framework is proposed for speaker embedding extraction [23].…”

Section: Introductionmentioning

confidence: 99%

The INTERSPEECH 2020 Far-Field Speaker Verification Challenge

Qin

et al. 2020

Interspeech 2020

View full text Add to dashboard Cite

The INTERSPEECH 2020 Far-Field Speaker Verification Challenge (FFSVC 2020) addresses three different research problems under well-defined conditions: far-field text-dependent speaker verification from single microphone array, far-field textindependent speaker verification from single microphone array, and far-field text-dependent speaker verification from distributed microphone arrays. All three tasks pose a cross-channel challenge to the participants. To simulate the real-life scenario, the enrollment utterances are recorded from close-talk cellphone, while the test utterances are recorded from the far-field microphone arrays. In this paper, we describe the database, the challenge, and the baseline system, which is based on a ResNetbased deep speaker network with cosine similarity scoring. For a given utterance, the speaker embeddings of different channels are equally averaged as the final embedding. The baseline system achieves minDCFs of 0.62, 0.66, and 0.64 and EERs of 6.27%, 6.55%, and 7.18% for task 1, task 2, and task 3, respectively.

show abstract

“…Notable works include masking [1] and mapping [2] based approach, Speech Enhancement Generative Adversarial Network (SEGAN) [3], Deep Feature Loss (DFL) [4], end-to-end metric optimization [5], and Transformer based approach [6,7]. Meanwhile, an active research exists in the robustness of Speaker Verification (SV) systems [8,9,10,11]. Another reason for interest in speech enhancement arises from the notion that it is considered as a modern solution to improve noise robustness in SV systems [10,12,13].…”

Section: Introductionmentioning

confidence: 99%

Analysis of Deep Feature Loss based Enhancement for Speaker Verification

Kataria¹,

Nidadavolu²,

Villalba³

et al. 2020

Preprint

View full text Add to dashboard Cite

Data augmentation is conventionally used to inject robustness in Speaker Verification systems. Several recently organized challenges focus on handling novel acoustic environments. Deep learning based speech enhancement is a modern solution for this. Recently, a study proposed to optimize the enhancement network in the activation space of a pre-trained auxiliary network. This methodology, called deep feature loss, greatly improved over the state-of-the-art conventional x-vector based system on a children speech dataset called BabyTrain. This work analyzes various facets of that approach and asks few novel questions in that context. We first search for optimal number of auxiliary network activations, training data, and enhancement feature dimension. Experiments reveal the importance of Signal-to-Noise Ratio filtering that we employ to create a large, clean, and naturalistic corpus for enhancement network training. To counter the "mismatch" problem in enhancement, we find enhancing front-end (x-vector network) data helpful while harmful for the back-end (Probabilistic Linear Discriminant Analysis (PLDA)). Importantly, we find enhanced signals contain complementary information to original. Established by combining them in front-end, this gives ~40% relative improvement over the baseline. We also do an ablation study to remove a noise class from x-vector data augmentation and, for such systems, we establish the utility of enhancement regardless of whether it has seen that noise class itself during training. Finally, we design several dereverberation schemes to conclude ineffectiveness of deep feature loss enhancement scheme for this task.

show abstract

Robust Speaker Recognition Using Speech Enhancement And Attention Model

Cited by 8 publications

References 34 publications

Dual Attention in Time and Frequency Domain for Voice Activity Detection

Dual Attention in Time and Frequency Domain for Voice Activity Detection

The INTERSPEECH 2020 Far-Field Speaker Verification Challenge

Analysis of Deep Feature Loss based Enhancement for Speaker Verification

Contact Info

Product

Resources

About