Filters








8,844 Hits in 5.6 sec

Speaking Style Authentication Using Suprasegmental Hidden Markov Models [article]

Ismail Shahin
2017 arXiv   pre-print
Our work in this research focuses on proposing, implementing, and testing speaker-dependent and text-dependent speaking style authentication (verification) systems that accept or reject the identity claim  ...  , shouted, slow, loud, soft, fast, angry, happy, and fearful.  ...  Multi-speaker training method requires a speaker to train an authentication system using speaking styles spoken by different speakers instead of using speaking styles all spoken by the same speaker.  ... 
arXiv:1706.09736v1 fatcat:cysdyfhpxzffthpnchqyn6kpey

PREFACE TO SPECIAL ISSUE ON BIOMETRICS AND APPLICATIONS

Marzuki Khalid
2009 International journal of biomedical soft computing and human sciences  
Next, Anas Quteishat et. al. proposed authenticating computer users through typing biometrics based on the fuzzy min-max neural approach. problem remains a major issue in many face recognition systems  ...  As pose their proposal is based on a of a set of feature computmg approach entitled DNA Computing fbr paper is not a typical strong potential in many biometric system applications.  ... 
doi:10.24466/ijbschs.14.1_1 fatcat:5odpdepktjg7zlhzxlfzi3igyy

Page 1194 of Linguistics and Language Behavior Abstracts: LLBA Vol. 23, Issue 3 [page]

1989 Linguistics and Language Behavior Abstracts: LLBA  
characteristics; 8904414 computer-assisted language learning systems, increased technological development possibilities; 8904562 computer-assisted language learning, natural language interfaces use; activities  ...  /future intelligent tutoring systems development suggested; 8904322 computer-assisted language learning, recent developments; 8904344 computer-assisted language teaching system, language acquisition model  ... 

Stride Towards Proposing Multi-Modal Biometric Authentication for Online Exam

A. Prakash, R. Dhanalakshmi
2016 International Journal of Network Security  
Also in the previous work they use face and dress color as hard and soft biometric traits.  ...  The major drawback of the existing continuous authentication system is, it is able to successfully authenticate the user continuously with high tolerance to the user posture.  ...  This framework registers a new en-rollment template every time the user logs in, which enables the system to effectively use soft biometric traits for continuous authentication.  ... 
dblp:journals/ijnsec/PrakashD16 fatcat:rjnonmcmqrhfneoan35yhqxh6a

Multimodal Biometric Person Authentication using Speech, Signature and Handwriting Features

Eshwarappa M, Mrityunjaya V.
2011 International Journal of Advanced Computer Science and Applications  
The objective of this work is to develop a multimodal biometric system using speech, signature and handwriting information.  ...  The latest research indicates multimodal person authentication system is more effective and more challenging.  ...  techniques, and Section III deals with multimodal biometric person authentication system by combining speaker, signature and handwriting recognition systems using fusion strategy.  ... 
doi:10.14569/specialissue.2011.010313 fatcat:6ngtfppkynf75non5zcmyoevia

Efficient approximated i-vector extraction

Hagai Aronowitz, Oren Barkan
2012 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
I-vectors are currently widely used by state-of-the-art speech processing systems for tasks such as speaker verification and language identification.  ...  Index Termsefficient speaker recognition, i-vectors, approximated i-vectors extraction 978-1-4673-0046-9/12/$26.00  ...  INTRODUCTION Recently, i-vector-based systems have become popular for speech processing systems such as speaker recognition [1] and language identification [2] .  ... 
doi:10.1109/icassp.2012.6288990 dblp:conf/icassp/AronowitzB12 fatcat:6hfsv57f5fgabiraalczcljzeu

LEAP System for SRE19 Challenge – Improvements and Error Analysis [article]

Shreyas Ramoji, Prashant Krishnan, Bhargavram Mysore, Prachi Singh, Sriram Ganapathy
2020 arXiv   pre-print
Following the x-vector extraction, we explored a neural network approach to backend score computation that was optimized for a speaker verification cost.  ...  All the systems used the time-delay neural network (TDNN) based x-vector embeddings. The x-vector system in our SRE19 submission used a large pool of training speakers (about 14k speakers).  ...  Introduction The recent years have seen increasing demand for authentication and verification systems using speech.  ... 
arXiv:2002.02735v1 fatcat:b7yhnjwvirah5hhtwnnyqylhcu

Soft Biometrics Traits for Continuous Authentication in Online Exam Using ICA Based Facial Recognition

Prakash Annamalai, Krishnaveni Raju, Dhanalakshmi Ranganayakulu
2018 International Journal of Network Security  
Continuous Authentication (CA) system verifies the user continuously once a person is logged in. Continuous Authentication system prevents the intruders from invoking the system.  ...  A new framework for continuous user authentication that primarily uses hard and soft biometric traits using Independent Components Analysis (ICA) dimension reducing method for video frames.  ...  The system continuously authenticates the user by using the "soft face" and "clothing" enrollment templates registered in initial login authentication.  ... 
dblp:journals/ijnsec/AnnamalaiRR18 fatcat:f4razrl63bg7fpumokddix3bc4

An Overview of Statistical Pattern Recognition Techniques for Speaker Verification

Amin Fazel, Shantanu Chakrabartty
2011 IEEE Circuits and Systems Magazine  
sample is authenticated using the trained speaker specific model.  ...  Authentication The authentication module uses the integrated likelihood scores to determine if the utterance belongs to the target speaker or belongs to an imposter.  ...  His current research interests include energy scavenging sensors and integrated circuits, hybrid circuits and systems and ultra-low power analog signal processing circuits. Dr.  ... 
doi:10.1109/mcas.2011.941080 fatcat:774dvpu63nevza5ii27ncw5nyu

Adversarial Disentanglement of Speaker Representation for Attribute-Driven Privacy Preservation [article]

Paul-Gauthier Noé, Mohammad Mohammadamini, Driss Matrouf, Titouan Parcollet, Andreas Nautsch, Jean-François Bonastre
2021 arXiv   pre-print
In speech technologies, speaker's voice representation is used in many applications such as speech recognition, voice conversion, speech synthesis and, obviously, user authentication.  ...  As a first solution to this concept, we propose to use an adversarial autoencoding method that disentangles in the voice representation a given speaker attribute thus allowing its concealment.  ...  The pre-trained sex classifier is used as a feature extractor that predicts the posteriorsỹ as soft labels.  ... 
arXiv:2012.04454v3 fatcat:axtcjx7qxngxlhuculnbt43gi4

A Comparison of Discrete and Soft Speech Units for Improved Voice Conversion [article]

Benjamin van Niekerk, Marc-André Carbonneau, Julian Zaïdi, Mathew Baas, Hugo Seuté, Herman Kamper
2021 arXiv   pre-print
We find that discrete representations effectively remove speaker information but discard some linguistic content - leading to mispronunciations. As a solution, we propose soft speech units.  ...  Samples available at https://ubisoft-laforge.github.io/speech/soft-vc/  ...  Then we add an equal number of authentic target speaker pairs.  ... 
arXiv:2111.02392v1 fatcat:aisilct3sza2necqmaxrjg5ale

Phonetic-attention scoring for deep speaker features in speaker verification [article]

Lantian Li, Zhiyuan Tang, Ying Shi, Dong Wang
2018 arXiv   pre-print
By this approach, an attention score is computed for each frame pair.  ...  We can use the same idea to align utterances with different phonetic contents. This paper proposes a phonetic-attention scoring approach for d-vector systems.  ...  This approach uses frame-level phonetic information to produce a soft alignment between the enrollment and test utterances, and computes the matching score by emphasizing the aligned frame pairs.  ... 
arXiv:1811.03255v1 fatcat:ry7rx7eiwrhlhi7o5bi4ebjwoi

Speaker Verification Using Deep Neural Networks: A Review

Amna Irum, School of Electrical Engineering and Computer Sciences (SEECS), National University of Sciences and Technology (NUST), Islamabad, Pakistan, Ahmad Salman
2019 International Journal of Machine Learning and Computing  
DNN are used from extracting features to complete end-to-end system for speaker verification.  ...  Speaker verification involves examining the speech signal to authenticate the claim of a speaker as true or false.  ...  Speaker recognition system works on the principle that every speaker's voice is unique like finger prints thus can be used to identify the speaker or authenticate his/her claim.  ... 
doi:10.18178/ijmlc.2019.9.1.760 fatcat:dskecbzey5eyhak5zv7wzq4eyq

Page 496 of Linguistics and Language Behavior Abstracts: LLBA Vol. 27, Issue 1 [page]

1993 Linguistics and Language Behavior Abstracts: LLBA  
imprecise computation technique, Sinhalese example; 9301758 natural language processing system INFANT, connectionist-like knowledge base; 9301740 natural language processing, connectionist finite state  ...  ; 9302390 Natural Language attribute/lexical-functional grammars, similarities/uses, computational theory perspective; 9301523 automatic natural language meaning interpretation, partial syntactic structure  ... 

Page 1606 of Linguistics and Language Behavior Abstracts: LLBA Vol. 26, Issue 3 [page]

1992 Linguistics and Language Behavior Abstracts: LLBA  
, language-independent grammatical error reporter system, artificial intelligence basis; 9205397 computer-assisted second-language instruction, text editing, ASOLI- YA/LINAR systems; 9205325 computer-based  ...  ; 9205346 computer-assisted French as a second language instruction, GEF soft- ware presentation; 9205177 computer-assisted German as a second language instruction, 1991 Leipzig colloquium; 9205355 Computer-Assisted  ... 
« Previous Showing results 1 — 15 out of 8,844 results