menu_book Explore the article's raw data

Empowering Speaker Verification with Deep Convolutional Neural Network Vectors

Abstract

This paper introduces a novel method for speaker verification using Convolutional Neural Networks (CNNs). Unlike traditional approaches that rely solely on spectrogram and waveform images, the proposed method, termed 'DeepConvVectors', dynamically captures speaker-specific features from speech signals. By transforming segments of speech into specialized CNN filters, Deep-ConvVectors were created, which encapsulate essential speaker characteristics. The experiments carried out on the THUYG-20 SRE dataset demonstrated the superior performance of the proposed method in comparison with the established methods, with an average Equal Error Rate (EER) of just 0.99%. This approach offers a dynamic solution for precise speaker identification, showcasing the transformative potential of CNNs in the context of ASV.

article Article

date_range 2024

language English

link Link of the paper

format_quote

Hourri Soufiane

Sorry! There is no raw data available for this article.

Loading references...

Loading citations...

Featured Keywords

Speaker verification

CNN

RBM

DBN

DNN

shopping_basket PURCHASE DATA

Assessment of reinforced concrete building damages following the Kahramanmaraş earthquakes in Malatya, Turkey (February 6, 2023)

Abstract

Earthquakes of magnitude Mw = 7.7 and Mw = 7.6 hit the city of Kahramanmaraş (Turkey) on 6 February 2023. These earthquakes caused serious damage in 11 provinces in Turkey resulting in over 50,000 deaths and huge economic losses with the collapse of thousands of buildings. Two consecutive devastating earthquakes have severely damaged the reinforced concrete building stock located in Malatya. There are many reasons reinforced concrete structures suffer damage under the influence of seismic loads. These reasons can be listed as material and section properties that do not comply with earthquake codes, workmanship defects, ground parameters that are not considered and design errors, as seen in field investigations after earthquakes in Turkey in the past. In this research, a field study was conducted on reinforced concrete buildings damaged after the earthquakes in Malatya province. The causes of structural damages in reinforced concrete buildings between 6–26 years of age were assessed in terms of design and material defects. The structural damages were mainly caused due to insufficient stirrups in structural members, short column, strong beam–weak column, failures of infill walls, soft story type collapse, poor concrete quality, and reinforcement corrosion. In addition to the field investigation, the spectrum relationships created using raw acceleration data obtained from DEMP (Disaster and Emergency Management Presidency) stations in Malatya province were compared with the design response spectra.

article Article

date_range 2024

language English

link Link of the paper

Collapsed buildings

Example 1 (jpg)