A Low-Power Text-Dependent Speaker Verification System with Narrow-Band Feature Pre-Selection and Weighted Dynamic Time Warping

Abstract In this paper, a brief summary of the author’s research in the field of the contour-based telephone speech Endpoint Detection (ED) is presented. This research includes: development of new robust features for ED – the Mean-Delta feature and the Group Delay Mean-Delta feature and estimation of the effect of the analyzed ED features and two additional features in the Dynamic Time Warping fixed-text speaker verification task with short noisy telephone phrases in Bulgarian language.

Download Full-text

Telephone Speech Endpoint Detection using Mean-Delta Feature

Cybernetics and Information Technologies ◽

10.2478/cait-2014-0025 ◽

2014 ◽

Vol 14 (2) ◽

pp. 127-139 ◽

Cited By ~ 3

Author(s):

Atanas Ouzounov

Keyword(s):

Dynamic Time Warping ◽

Speaker Verification ◽

Verification Task ◽

Endpoint Detection ◽

Time Warping ◽

Energy Entropy ◽

Telephone Speech ◽

Dynamic Time ◽

Teager Energy ◽

Speech Endpoint Detection

Abstract In the study the efficiency of three features for trajectory-based endpoint detection is experimentally evaluated in the fixed-text Dynamic Time Warping (DTW) - a based speaker verification task with short phrases of telephone speech. The employed features are Modified Teager Energy (MTE), Energy-Entropy (EE) feature and Mean-Delta (MD) feature. The utterance boundaries in the endpoint detector are provided by means of state automaton and a set of thresholds based only on trajectory characteristics. The training and testing have been done with noisy telephone speech (short phrases in Bulgarian language with length of about 2 s) selected from BG-SRDat corpus. The results of the experiments have shown that the MD feature demonstrates the best performance in the endpoint detection tests in terms of the verification rate.

Download Full-text

K-Nearest Neighbour and Dynamic Time Warping for Online Signature Verification

10.5121/csit.2021.111813 ◽

2021 ◽

Author(s):

Mohammad Saleem ◽

BenceKovari

Keyword(s):

Error Rate ◽

Dynamic Time Warping ◽

Signature Verification ◽

Nearest Neighbour ◽

Time Warping ◽

Verification System ◽

Online Signature ◽

Online Signature Verification ◽

Dynamic Time ◽

Verification Systems

Online signatures are one of the most commonly used biometrics. Several verification systems and public databases were presented in this field. This paper presents a combination of knearest neighbor and dynamic time warping algorithms as a verification system using the recently published DeepSignDB database. Our algorithm was applied on both finger and stylus input signatures which represent both office and mobile scenarios. The system was first tested on the development set of the database. It achieved an error rate of 6.04% for the stylus input signatures, 5.20% for the finger input signatures, and 6.00% for a combination of both types. The system was also applied to the evaluation set of the database and achieved very promising results, especially for finger input signatures.

Download Full-text

Addressing Text-Dependent Speaker Verification Using Singing Speech

Applied Sciences ◽

10.3390/app9132636 ◽

2019 ◽

Vol 9 (13) ◽

pp. 2636 ◽

Cited By ~ 2

Author(s):

Yan Shi ◽

Juanjuan Zhou ◽

Yanhua Long ◽

Yijie Li ◽

Hongwei Mao

Keyword(s):

Dynamic Time Warping ◽

State Of The Art ◽

Speaker Verification ◽

Gaussian Mixture ◽

Time Warping ◽

Normal Reading ◽

Feature Spaces ◽

Dynamic Time ◽

The One ◽

Normal Speech

The automatic speaker verification (ASV) has achieved significant progress in recent years. However, it is still very challenging to generalize the ASV technologies to new, unknown and spoofing conditions. Most previous studies focused on extracting the speaker information from natural speech. This paper attempts to address the speaker verification from another perspective. The speaker identity information was exploited from singing speech. We first designed and released a new corpus for speaker verification based on singing and normal reading speech. Then, the speaker discrimination was compared and analyzed between natural and singing speech in different feature spaces. Furthermore, the conventional Gaussian mixture model, the dynamic time warping and the state-of-the-art deep neural network were investigated. They were used to build text-dependent ASV systems with different training-test conditions. Experimental results show that the voiceprint information in the singing speech was more distinguishable than the one in the normal speech. More than relative 20% reduction of equal error rate was obtained on both the gender-dependent and independent 1 s-1 s evaluation tasks.

Download Full-text

Building Sequence Kernels for Speaker Verification and Word Recognition

Intelligent Information Technologies ◽

10.4018/978-1-59904-941-0.ch033 ◽

2011 ◽

pp. 575-589

Author(s):

Vincent Wan

Keyword(s):

Speech Recognition ◽

Speech Processing ◽

Kernel Methods ◽

Speaker Recognition ◽

Dynamic Time Warping ◽

Speaker Verification ◽

Dimensional Space ◽

Time Warping ◽

Recognition Systems ◽

Dynamic Time

This chapter describes the adaptation and application of kernel methods for speech processing. It is divided into two sections dealing with speaker verification and isolated-word speech recognition applications. Significant advances in kernel methods have been realised in the field of speaker verification, particularly relating to the direct scoring of variable-length speech utterances by sequence kernel SVMs. The improvements are so substantial that most state-of-the-art speaker recognition systems now incorporate SVMs. We describe the architecture of some of these sequence kernels. Speech recognition presents additional challenges to kernel methods and their application in this area is not as straightforward as for speaker verification. We describe a sequence kernel that uses dynamic time warping to capture temporal information within the kernel directly. The formulation also extends the standard dynamic time-warping algorithm by enabling the dynamic alignment to be computed in a high-dimensional space induced by a kernel function. This kernel is shown to work well in an application for recognising low-intelligibility speech of severely dysarthric individuals.

Download Full-text

Building Sequence Kernels for Speaker Verification and Word Recognition

Kernel Methods in Bioengineering, Signal and Image Processing ◽

10.4018/978-1-59904-042-4.ch010 ◽

2011 ◽

pp. 246-262

Author(s):

Vincent Wan

Keyword(s):

Speech Recognition ◽

Speech Processing ◽

Kernel Methods ◽

Speaker Recognition ◽

Dynamic Time Warping ◽

Speaker Verification ◽

Dimensional Space ◽

Time Warping ◽

Recognition Systems ◽

Dynamic Time

This chapter describes the adaptation and application of kernel methods for speech processing. It is divided into two sections dealing with speaker verification and isolated-word speech recognition applications. Significant advances in kernel methods have been realised in the field of speaker verification, particularly relating to the direct scoring of variable-length speech utterances by sequence kernel SVMs. The improvements are so substantial that most state-of-the-art speaker recognition systems now incorporate SVMs. We describe the architecture of some of these sequence kernels. Speech recognition presents additional challenges to kernel methods and their application in this area is not as straightforward as for speaker verification. We describe a sequence kernel that uses dynamic time warping to capture temporal information within the kernel directly. The formulation also extends the standard dynamic time-warping algorithm by enabling the dynamic alignment to be computed in a high-dimensional space induced by a kernel function. This kernel is shown to work well in an application for recognising low-intelligibility speech of severely dysarthric individuals.

Download Full-text

LTSD and GDMD features for Telephone Speech Endpoint Detection

Cybernetics and Information Technologies ◽

10.1515/cait-2017-0045 ◽

2017 ◽

Vol 17 (4) ◽

pp. 114-133

Author(s):

Atanas Ouzounov

Keyword(s):

Dynamic Time Warping ◽

Group Delay ◽

Speaker Verification ◽

Endpoint Detection ◽

Time Warping ◽

Adaptive Thresholds ◽

Telephone Speech ◽

Dynamic Time ◽

Speech Endpoint Detection

AbstractThis paper proposes a new contour-based speech endpoint detector which combines the log-Group Delay Mean-Delta (log-GDMD) feature, an adaptive twothreshold scheme and an eight-state automaton. The adaptive thresholds scheme uses two pairs of thresholds - for the starting and for the ending points, respectively. Each pair of thresholds is calculated by using the contour characteristics in the corresponded region of the utterance. The experimental results have shown that the proposed detector demonstrates better performance compared to the Long-Term Spectral Divergence (LTSD) one in terms of endpoint accuracy. Additional fixed-text speaker verification tests with short phrases of telephone speech based on the Dynamic Time Warping (DTW) and left-to-right Hidden Markov Model (HMM) frameworks confirm the improvements of the verification rate due to the better endpoint accuracy.

Download Full-text