Filters








2,698 Hits in 7.3 sec

Extraction of templates from phrases using Sequence Binary Decision Diagrams

D. HIRANO, K. TANAKA-ISHII, A. FINCH
2018 Natural Language Engineering  
This paper presents an unsupervised approach for extracting templates on-the-fly from only tagged text by using a novel relaxed variant of the Sequence Binary Decision Diagram (SeqBDD).  ...  AbstractThe extraction of templates such as 'regard X as Y' from a set of related phrases requires the identification of their internal structures.  ...  Among the different data structures that allow this type of sharing, we chose the SeqBDD (Sequence Binary Decision Diagram) as our basis.  ... 
doi:10.1017/s1351324918000268 fatcat:2l6wrxz4wjcrjgvuvb5vyrck3m

Voice-activated word processor with automatic learning for dynamic optimization of syllable-templates

Furnio Togawa, Mitsuhiro Hakaridani, Hiroyuki Iwahashi, Torn Ueda
1989 Journal of the Acoustical Society of Japan (E)  
The automatic learning algorithm replaces low-accuracy syllable templates with new patterns extracted from the input speech.  ...  Such learning information is derived from comparison between syllable recognition results and a user-confirmed character string of input phrase as correct, and is accumulated on a phrase-by-phrase basis  ...  The matrix is a three-dimensional binary matrix and specifies whether a specific sequence of three syllables can occur within a phrase.  ... 
doi:10.1250/ast.10.133 fatcat:ykyi76csvbhhjlcebljfr7jzge

Secure speech biometric templates for user authentication

K. Inthavisas, D. Lopresti
2012 IET Biometrics  
A set of pass-phrases from the speakers is used to create templates of M classes. Each class consists of R reference templates.  ...  If a binary string generated from the hardened template is random, the expected agreement between the binary template and the binary string derived from the hardened template should be close to 256 bits  ...  Multi-threshold template .  ... 
doi:10.1049/iet-bmt.2011.0008 fatcat:jfcwognuazaa7nfwq6hisjayj4

Steerable Music Generation which Satisfies Long-Range Dependency Constraints

Paul Bodily, Dan Ventura
2022 Transactions of the International Society for Music Information Retrieval  
We present RegulaR nonhomogeneous Markov models as a solution to the long-range dependency problem which uses Relational automata to enforce binary constraints to compose music with repeating motifs.  ...  Although music is full of repetitive motifs and themes, artificially intelligent temporal sequence models have yet to demonstrate the ability to model or generate musical compositions that satisfy steerable  ...  Multivalued Decision Diagrams (MDDs) can be used to model a combination of Markov and regular constraints (Perez and Régin, 2017) .  ... 
doi:10.5334/tismir.97 fatcat:w4xjjrhylzdodkifdhqsneqm4q

A Template-Based Approach for Speech Synthesis Intonation Generation Using LSTMs

Srikanth Ronanki, Gustav Eje Henter, Zhizheng Wu, Simon King
2016 Interspeech 2016  
The use of syllable templates mitigates the over-smoothing problem and is able to reproduce pitch patterns observed in the data.  ...  In order to capture and reproduce larger-scale pitch patterns, this paper proposes a template-based approach for automatic F0 generation, where per-syllable pitchcontour templates (from a small, automatically  ...  class from input text features; 3) F0 contour reconstruction from this sequence of templates.  ... 
doi:10.21437/interspeech.2016-96 dblp:conf/interspeech/RonankiHWK16 fatcat:ae74ejo36ffyzbdtfsgy4j5oce

Deep Template Matching for Small-Footprint and Configurable Keyword Spotting

Peng Zhang, Xueliang Zhang
2020 Interspeech 2020  
Keyword spotting (KWS) is a very important technique for human-machine interaction to detect a trigger phrase and voice commands.  ...  In this paper, we propose a novel template matching approach for KWS based on end-to-end deep learning method, which utilizes an attention mechanism to match the input voice to the keyword templates in  ...  vector, f , through the template matching module, we use a binary classifier with cross-entropy loss to make the final decision for the input pair.  ... 
doi:10.21437/interspeech.2020-1761 dblp:conf/interspeech/ZhangZ20 fatcat:5q2fk57e3vbvrjqydakdietle4

A Review on Speaker Recognition

Sujiya S, Dr.Chandra E
2017 International Journal of Engineering and Technology  
This paper depicts the fundamentals of automatic speaker recognition, concerning feature extraction and speaker modeling.  ...  Today many employees have access to their company's information system by logging in from home. Also Internet services and telephone banking are widely used by private and corporate sectors.  ...  This is done using sequences of feature vectors extracted from multiple utterances of the same word by the same person.  ... 
doi:10.21817/ijet/2017/v9i3/170903513 fatcat:z3pulpp65vaqdmtjlqh75dd52y

Lipreading With Local Spatiotemporal Descriptors

Guoying Zhao, M. Barnard, M. Pietikainen
2009 IEEE transactions on multimedia  
Spatiotemporal local binary patterns extracted from mouth regions are used for describing isolated phrase sequences.  ...  In our experiments with 817 sequences from ten phrases and 20 speakers, promising accuracies of 62% and 70% were obtained in speaker-independent and speaker-dependent recognition, respectively.  ...  Kontinen for helping collect the OuluVS database used in the paper and Dr. R. Harvey for providing the AVLetters database used in experiments.  ... 
doi:10.1109/tmm.2009.2030637 fatcat:n5dqgpwezjebzo3ji6y5i5joby

A complete traceability methodology between UML diagrams and source code based on enriched use case textual description

Wiem Khlif, Dhikra Kchaou, Nadia Bouassida
2022 Informatica (Ljubljana, Tiskana izd.)  
In fact, the extension consists in extracting an expanded textual description from a natural language text in order to trace between related elements belonging to requirement, design and code while using  ...  The finetuning of our methodology stems from considering an expanded textual description.  ...  NLPC uses pre-processing, Part of Speech (POS) Tagging, Class Identification, Attribute and Function identification to plot the classes. [5] extracts class diagrams from natural language requirements using  ... 
doi:10.31449/inf.v46i1.3306 dblp:journals/informaticaSI/KhlifKB22 fatcat:futxzetwxbcppcdg2zcg755mki

Comparative Study of Biometric Models for Individuality Investigation

Oluwatayo Samuel Ogunlana
2021 International Journal of Computer Applications  
Some of these problems can be overcome by integrating the use of multiple biometric traits of a user (e.g. face, fingerprint).  ...  In the entire world, security systems are essential for the protection of life and property. This is a growing technology which has become increasingly used in our daily life.  ...  In motion based approach, the human gait is considered as a sequence of image and features are extracted from these images while, the model based approach extract the motion of the human body by means  ... 
doi:10.5120/ijca2021921540 fatcat:luh63eno5baezas4r5acudqcnu

A highly efficient system for automatic face region detection in MPEG video

Hualu Wang, Shih-Fu Chang
1997 IEEE transactions on circuits and systems for video technology (Print)  
Human faces provide a useful cue in indexing video content. In this paper, we present a highly efficient system that can rapidly detect human face regions in MPEG video sequences.  ...  The algorithm consists of three stages, where chrominance, shape, and frequency information are used respectively.  ...  Harold Stone of NEC for his valuable comments. We also thank CNN for the permission to use the test video in our experiments.  ... 
doi:10.1109/76.611173 fatcat:nuncvx5zevetfiquinxmaelnsa

Event Schema Induction with a Probabilistic Entity-Driven Model

Nathanael Chambers
2013 Conference on Empirical Methods in Natural Language Processing  
Event schema induction is the task of learning high-level representations of complex events (e.g., a bombing) and their entity roles (e.g., perpetrator and victim) from unlabeled text.  ...  Event schemas have important connections to early NLP research on frames and scripts, as well as modern applications like template extraction.  ...  Acknowledgments This work was partially supported by a grant from the Office of Naval Research. It was also supported, in part, by the Johns Hopkins Human Language Technology Center of Excellence.  ... 
dblp:conf/emnlp/Chambers13 fatcat:wnd6hwvb4vh2ppb5kr425olkgy

A Survey on Biometrics and Cancelable Biometrics Systems

Bismita Choudhury, Patrick Then, Biju Issac, Valliappan Raman, Manas Kumar Haldar
2018 International Journal of Image and Graphics  
Ear Geometry The shape of the outer ear, bone structure and ear lobe are considered for person recognition using the ear.  ...  To deal with the issue of the compromised biometric template, template protection schemes evolved to make it possible to replace the biometric template.  ...  The original biometric pattern is characterised as discrete sequence. From the original sequence, a set of sequence is extracted.  ... 
doi:10.1142/s0219467818500067 fatcat:nvjolkcjxbertlxmti3phlksw4

Cascaded Transforms for Signature Recognition with Image Registration

M. ManiRoja, Sudhir Sawarkar
2012 International Journal of Computer Applications  
The signatures are acquired using WACOM digital pen tablet and then features are extracted.  ...  Our experimental results on the image data set from 75 users, confirm the advantages of our cascaded approach over unimodal approaches.  ...  FEATURE EXTRACTION To extract the feature vectors from the signature template, we have used the following transforms. .Discrete Wavelet Transform The Wavelet transform [8] [9] [10] is a transform which  ... 
doi:10.5120/4795-7046 fatcat:hsfszwhx5jabrpfu3js5gjisda

Text Chunking Using Transformation-Based Learning [chapter]

L. A. Ramshaw, M. P. Marcus
1999 Text, Speech and Language Technology  
In automatic tests using Treebank-derived data, this technique achieved recall and precision rates of roughly 92% for baseNP chunks and 88% for somewhat more complex chunks that partition the sentence.  ...  The same method can be applied at a higher level of textual interpretation for locating chunks in the tagged text, including non-recursive "baseNP" chunks.  ...  . aNote that this is one of the cases where Church's chunker allows separate NP fragments to count as chunks.  ... 
doi:10.1007/978-94-017-2390-9_10 fatcat:pifs2to2lfcatoxzdg6ftyzzqq
« Previous Showing results 1 — 15 out of 2,698 results