Fast training of Large Margin diagonal Gaussian mixture models for speaker identification

机译：扬声器识别的大型裕度对角线高斯混合模型的快速训练

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Gaussian mixture models (GMM) have been widely and successfully used in speaker recognition during the last decades. They are generally trained using the generative criterion of maximum likelihood estimation. In an earlier work, we proposed an algorithm for discriminative training of GMM with diagonal covariances under a large margin criterion. In this paper, we present a new version of this algorithm which has the major advantage of being computationally highly efficient. The resulting algorithm is thus well suited to handle large scale databases. We carry out experiments on a speaker identification task using NIST-SRE'2006 data and compare our new algorithm to the baseline generative GMM using different GMM sizes. The results show that our system significantly outperforms the baseline GMM in all configurations, and with high computational efficiency.

机译：高斯混合模型（GMM）已广泛且成功地在过去几十年中成功地用于扬声器识别。通常使用最大似然估计的生成标准训练它们。在早期的工作中，我们提出了一种在大幅标准下具有对角线协方差的GMM的判别训练算法。在本文中，我们提出了一种新版本的该算法，其具有计算上高效的主要优点。因此，所得到的算法非常适合处理大规模数据库。我们使用NIST-SRE'2006数据对扬声器识别任务进行实验，并将我们的新算法与基线生成GMM进行比较，使用不同的GMM尺寸。结果表明，我们的系统在所有配置中显着优于基线GMM，并具有高计算效率。

著录项

来源
《Conference on Speech Technology and Human-Computer Dialogue》|2011年||共4页
会议地点
作者
Jourani Reda; Daoudi Khalid; Andre-Obrecht Regine; Aboutajdine Driss;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类 TN912-53;
关键词
Gaussian mixture models; discriminative learning; large margin training; speaker identification; speaker recognition;

机译：高斯混合模型;歧视学习;大幅度训练;扬声器识别;发言人识别;

相似文献

外文文献
中文文献
专利

1. Robust text-independent speaker identification using Gaussian mixture speaker models [J] . Reynolds D.A., Rose R.C. IEEE Transactions on Speech and Audio Proceeding . 1995,第1期

机译：使用高斯混合说话人模型进行鲁棒的与文本无关的说话人识别
2. Calculating Model Parameters Using Gaussian Mixture Models Based on Vector Quantization in Speaker Identification [J] . Hamideh Rezaei-Nezhad International journal of computer science and network security . 2017,第2期

机译：基于矢量量化的高斯混合模型在说话人识别中的模型参数计算
3. Analysis And Identification Of Emotion Specific Features For Speaker Independent Emotion Recognition System Using Gaussian Mixture Models (GMMs) [J] . J. Naga Padmaja, R. RajeswarRao Advances in computational sciences and technology . 2017,第8PTa2343a2506期

机译：基于高斯混合模型（GMM）的独立于说话人的情绪识别系统的情绪特定特征的分析和识别
4. Fast training of Large Margin diagonal Gaussian mixture models for speaker identification [C] . Jourani Reda, Daoudi Khalid, Andre-Obrecht Regine, Proceedings of the 6th International Conference on Speech Technology and Human-Computer Dialogue . 2011

机译：快速训练大余量对角线高斯混合模型以识别说话人
5. A software based speaker identification system using Gaussian mixture model classification. [D] . Reynolds, Ryan M. 2005

机译：使用高斯混合模型分类的基于软件的说话人识别系统。
6. Segway 2.0: Gaussian mixture models and minibatch training [O] . Rachel C W Chan, Maxwell W Libbrecht, Eric G Roberts, -1

机译：Segway 2.0：高斯混合模型和小批量训练
7. Metodología de entrenamiento de modelos de mezclas gaussianas empleando criterios de gran margen para la detección de patologías en bioseñales = Training Methodology of Gaussian Mixture Models byudEmploying Large Margin to Detect Pathologies in Biosignals [O] . Carvajal González Johanna Paola 2010

机译：使用大边缘标准检测生物信号中病理学的高斯混合模型的训练方法=高斯混合模型的训练方法利用大边距检测生物信号中的病理

Fast training of Large Margin diagonal Gaussian mixture models for speaker identification

摘要

著录项

相似文献

相关主题

期刊订阅