MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION
The paper proposes using fuzzy integrals for merging classifier decisions in speaker recognitionsystems. Instantaneous frequency and instantaneous amplitude are considered as the set of features. The approach shows significantly better results than a single classifier. A comparison of the proposed a...
Main Authors: | , |
---|---|
Format: | Article |
Language: | Russian |
Published: |
The United Institute of Informatics Problems of the National Academy of Sciences of Belarus
2016-09-01
|
Series: | Informatika |
Online Access: | https://inf.grid.by/jour/article/view/3 |
_version_ | 1797877401879838720 |
---|---|
author | Y. N. Imamverdiyev L. V. Sukhostat |
author_facet | Y. N. Imamverdiyev L. V. Sukhostat |
author_sort | Y. N. Imamverdiyev |
collection | DOAJ |
description | The paper proposes using fuzzy integrals for merging classifier decisions in speaker recognitionsystems. Instantaneous frequency and instantaneous amplitude are considered as the set of features. The approach shows significantly better results than a single classifier. A comparison of the proposed approach with the other methods for merging classifier decisions is provided. |
first_indexed | 2024-04-10T02:17:44Z |
format | Article |
id | doaj.art-bed4afcaf9734e27b685d535c5429d92 |
institution | Directory Open Access Journal |
issn | 1816-0301 |
language | Russian |
last_indexed | 2024-04-10T02:17:44Z |
publishDate | 2016-09-01 |
publisher | The United Institute of Informatics Problems of the National Academy of Sciences of Belarus |
record_format | Article |
series | Informatika |
spelling | doaj.art-bed4afcaf9734e27b685d535c5429d922023-03-13T08:32:16ZrusThe United Institute of Informatics Problems of the National Academy of Sciences of BelarusInformatika1816-03012016-09-010117252MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITIONY. N. Imamverdiyev0L. V. Sukhostat1Институт информационных технологий Национальной академии наук АзербайджанаИнститут информационных технологий Национальной академии наук АзербайджанаThe paper proposes using fuzzy integrals for merging classifier decisions in speaker recognitionsystems. Instantaneous frequency and instantaneous amplitude are considered as the set of features. The approach shows significantly better results than a single classifier. A comparison of the proposed approach with the other methods for merging classifier decisions is provided.https://inf.grid.by/jour/article/view/3 |
spellingShingle | Y. N. Imamverdiyev L. V. Sukhostat MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION Informatika |
title | MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION |
title_full | MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION |
title_fullStr | MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION |
title_full_unstemmed | MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION |
title_short | MERGING CLASSIFIER DECISIONS FOR SPEAKER RECOGNITION |
title_sort | merging classifier decisions for speaker recognition |
url | https://inf.grid.by/jour/article/view/3 |
work_keys_str_mv | AT ynimamverdiyev mergingclassifierdecisionsforspeakerrecognition AT lvsukhostat mergingclassifierdecisionsforspeakerrecognition |