A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments

Jung, Yongju; Choi, Yeunju; Lim, Hyungjun; Kim, Hoirin

doi:10.1109/access.2020.3025941

Cited by 17 publications

(5 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We utilize the power of Convolutional Neural Networks (CNNs) and technology derived from the VGG-Net model [3] to further optimize our lightweight model. These advancements allow us to identify distinct speech patterns, increasing system accuracy and lowering resource needs at the same time.…”

Section: Introductionmentioning

confidence: 99%

A Lightweight Approach Towards Speaker Authentication Systems

More,

Patil,

Deshmukh

et al. 2023

Preprint

View full text Add to dashboard Cite

In a world where traditional authentication systems are constrained, the introduction of voice-based authentication provides a viable option. This cutting-edge biometric security method uses unique vocal traits including pitch, tone, and speech patterns to create a unique voiceprint. Its sophisticated and dependable verification procedure, when combined with voice-activated services and secure access systems, not only solves au-thentication issues but also improves user experience overall. Our suggested method provides a thorough blueprint for developing a lightweight voice authentication system that is based on text. This system leverages an effective encoder model to convert voice spectrograms, making deployment easy even on edge devices with limited resources. Various dimensionality reduction techniques are explored to obtain optimal voice embeddings that capture speaker uniqueness while minimizing model complexity. A key novelty is the application of model compression techniques including lightweight architectures and Siamese Networks to obtain highly condensed voice embeddings for each user, reducing storage and infrastructure costs compared to traditional voice authentication methods. The proposed lightweight spectrogram embeddings leverage language-agnostic acoustic features, enabling language-independent speaker verification. Additionally, dimensionality reduction applied during voice registration allows the capturing of discriminative voice characteristics in a low-dimensional compact feature space. This significantly cuts down the storage requirements and infrastructure costs per user compared to standard voice biometric approaches, while retaining competitive verification performance. The architecture is optimized for responsiveness by leveraging lightweight frameworks. The proposed system delivers competitive voice authentication capabilities while minimizing memory, computational, and energy footprints. This makes the system useful for integration into smart devices and paves the way for ubiquitous voice biometrics.

show abstract

Section: Introductionmentioning

confidence: 99%

A Lightweight Approach Towards Speaker Authentication Systems

More,

Patil,

Deshmukh

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…THE goal of acoustic signal enhancement (ASE) is to suppress the interfering noise signals by minimizing unwanted distortions and transforming noisy input signals into desired clean signals. Acoustic signals are often distorted owing to additive and convolutional noise, or recording device constraints, which limit the performance of real-world applications, such as soundscape information retrieval [1][2][3], sound environment analysis in a smart home [4][5][6], physiological sound recognition [7][8][9][10], speaker recognition and verification [11][12][13][14], automatic speech recognition (ASR) [15][16][17][18], hearing aids [19,20], and cochlear implants [21,22]. Several ASE approaches have been proposed in the literature to alleviate background noise problems.…”

Section: Introductionmentioning

confidence: 99%

A Novel Temporal Attentive-Pooling based Convolutional Recurrent Architecture for Acoustic Signal Enhancement

Hussain

Wang

Gogate

et al. 2022

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

Impact Statement -Recently proposed deep learning solutions have proven useful in overcoming certain limitations of conventional acoustic signal enhancement (ASE) tasks. However, the performance of these approaches under real acoustic conditions is not always satisfactory. In this study, we investigated the use of attention models for ASE. To the best of our knowledge, this is the first attempt to successfully employ a convolutional recurrent neural network (CRNN) with a temporal attentive pooling (TAP) algorithm for the ASE task. The proposed TAP-CRNN framework can practically benefit the assistive communication technology industry, such as the manufacture of hearing aid devices for the elderly and students. In addition, the derived algorithm can benefit other signal processing applications, such as soundscape information retrieval, sound environment analysis in smart homes, and automatic speech/speaker/language recognition systems.

show abstract

“…For example, the presence of noise affecting human speech during the recording or the need for a very long speech signal to be recorded by the user to train the system or reverberation affecting the system. These challenges are discussed in [24], where the authors propose a method for short fragments of speech signals tackling the issues above described.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Mitigants for Speech Based Cyber Risk

Campi¹,

Peters

Azzaoui

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Statistical analysis of speech is an emerging area of machine learning. In this paper, we tackle the biometric challenge of Automatic Speaker Verification (ASV) of differentiating between samples generated by two distinct populations of utterances, those of an authentic human voice and those generated by a synthetic one. Solving such an issue through a statistical perspective foresees the definition of a decision rule function and a learning procedure to identify the optimal classifier. Classical state-ofthe-art countermeasures rely on strong assumptions such as stationarity or local-stationarity of speech that may be atypical to encounter in practice. We explore in this regard a robust non-linear and nonstationary signal decomposition method known as the Empirical Mode Decomposition combined with the Mel-Frequency Cepstral Coefficients in a novel fashion with a refined classifier technique known as multi-kernel Support Vector machine. We undertake significant real data case studies covering multiple ASV systems using different datasets, including the ASVSpoof 2019 challenge database. The obtained results overwhelmingly demonstrate the significance of our feature extraction and classifier approach versus existing conventional methods in reducing the threat of cyber-attack perpetrated by synthetic voice replication seeking unauthorised access.

show abstract

A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments

Cited by 17 publications

References 63 publications

A Lightweight Approach Towards Speaker Authentication Systems

A Lightweight Approach Towards Speaker Authentication Systems

A Novel Temporal Attentive-Pooling based Convolutional Recurrent Architecture for Acoustic Signal Enhancement

Machine Learning Mitigants for Speech Based Cyber Risk

Contact Info

Product

Resources

About