Speaker Authentication System Based on Voice Biometrics and Speech Recognition

Dovydaitis, Laurynas; Rasymas, Tomas; Rudzionis, Vytautas

doi:10.1007/978-3-319-52464-1_8

Cited by 6 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the unique and hard-forged characteristics of human biology, biometrics has emerged as a novel and robust technology in many verification tasks. At present, biometric verification methods based on fingerprint [ 22 ], face [ 23 ], and voice [ 24 ] have been used widely and proved to have relatively high accuracy. Yet, recent study showed that the above methods forged security risks [ 9 , 25 ].…”

Section: Background and Related Workmentioning

confidence: 99%

A Continuous Identity Authentication Scheme Based on Physiological and Behavioral Characteristics

Wang

Zhang

et al. 2018

Sensors

View full text Add to dashboard Cite

Wearable devices have flourished over the past ten years providing great advantages to people and, recently, they have also been used for identity authentication. Most of the authentication methods adopt a one-time authentication manner which cannot provide continuous certification. To address this issue, we present a two-step authentication method based on an own-built fingertip sensor device which can capture motion data (e.g., acceleration and angular velocity) and physiological data (e.g., a photoplethysmography (PPG) signal) simultaneously. When the device is worn on the user’s fingertip, it will automatically recognize whether the wearer is a legitimate user or not. More specifically, multisensor data is collected and analyzed to extract representative and intensive features. Then, human activity recognition is applied as the first step to enhance the practicability of the authentication system. After correctly discriminating the motion state, a one-class machine learning algorithm is applied for identity authentication as the second step. When a user wears the device, the authentication process is carried on automatically at set intervals. Analyses were conducted using data from 40 individuals across various operational scenarios. Extensive experiments were executed to examine the effectiveness of the proposed approach, which achieved an average accuracy rate of 98.5% and an F1-score of 86.67%. Our results suggest that the proposed scheme provides a feasible and practical solution for authentication.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

A Continuous Identity Authentication Scheme Based on Physiological and Behavioral Characteristics

Wang

Zhang

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…Nowadays, speaker verification has become a crucial task in many applications, such as security, authentication, surveillance, forensics, multi-speaker tracking, personalized user interfaces, voice assistants, door access control systems and others [4][5][6][7]. Nonetheless, the most important capability of speaker verification is the ability to identify and distinguish between different speakers accurately to be able to pinpoint a specific speaker.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Modern Deep Learning Models for Speaker Verification

Brydinskyi,

Khoma,

Sabodashko

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

This research presents an extensive comparative analysis of a selection of popular deep speaker embedding models, namely WavLM, TitaNet, ECAPA, and PyAnnote, applied in speaker verification tasks. The study employs a specially curated dataset, specifically designed to mirror the real-world operating conditions of voice models as accurately as possible. This dataset includes short, non-English statements gathered from interviews on a popular online video platform. The dataset features a wide range of speakers, with 33 males and 17 females, making a total of 50 unique voices. These speakers vary in age from 20 to 70 years old. This variety helps in thoroughly testing speaker verification models. This dataset is especially useful for research on speaker verification with short recordings. It consists of 10 clips for each person, each clip being no longer than 10 s, adding up to 500 recordings in total. The total length of all recordings is about 1 h and 30 min, which averages to roughly 100 s for each speaker. This dataset is a valuable tool for research in speaker verification, particularly for studies involving short audio clips. The performance of these models is evaluated using common biometric metrics such as false acceptance rate (FAR), false rejection rate (FRR), equal error rate (EER) and detection cost function (DCF). The results reveal that the TitaNet and ECAPA models stand out by presenting the lowest EER (1.91% and 1.71%, respectively) and thus exhibiting higher discriminative features, ensuring, on the one hand, a reduction in intra-class distance (the same speaker), and, on the other hand, maximizing the distance between different speaker embeddings. This analysis also highlights the ECAPA model’s advantageous balance of performance and efficiency, achieving an inference time of 69.43 milliseconds, slightly longer than the PyAnnote models. This study not only compares the performance of models but also provides a comparative analysis of respective model embeddings, offering insights into their strengths and weaknesses. The presented findings serve as a foundation for guiding future research in speaker verification, especially in the context of short audio samples or limited data. This may be particularly relevant for applications requiring quick and accurate speaker identification from short voice clips.

show abstract

“…Compared to the highlighted properties in related areas, such as the undetectability in steganography [4] and the amount capacity in data hiding [5], an image watermarking scheme often considers the robustness that ensures correct watermark extraction and recognition when the marked-image is distorted or disturbed. Audio watermarks: From the aspect of applications, an audio watermark facilitates different downstream scenarios by providing multiple information, for example, two-factor authentication [6], [7] by providing the speaker and content information. For research investigation, there is quite a number of work that separately focused on cover-image and cover-audio watermarking [8]- [10].…”

Section: Introductionmentioning

confidence: 99%

A Deep Learning-based Audio-in-Image Watermarking Scheme

Das

Zhong

2021

Preprint

View full text Add to dashboard Cite

This paper presents a deep learning-based audioin-image watermarking scheme. Audio-in-image watermarking is the process of covertly embedding and extracting audio watermarks on a cover-image. Using audio watermarks can open up possibilities for different downstream applications. For the purpose of implementing an audio-in-image watermarking that adapts to the demands of increasingly diverse situations, a neural network architecture is designed to automatically learn the watermarking process in an unsupervised manner. In addition, a similarity network is developed to recognize the audio watermarks under distortions, therefore providing robustness to the proposed method. Experimental results have shown high fidelity and robustness of the proposed blind audio-in-image watermarking scheme.

show abstract

Speaker Authentication System Based on Voice Biometrics and Speech Recognition

Cited by 6 publications

References 6 publications

A Continuous Identity Authentication Scheme Based on Physiological and Behavioral Characteristics

A Continuous Identity Authentication Scheme Based on Physiological and Behavioral Characteristics

Comparison of Modern Deep Learning Models for Speaker Verification

A Deep Learning-based Audio-in-Image Watermarking Scheme

Contact Info

Product

Resources

About