• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Arabian Journal for Science and Engineering. Section A, Sciences >Bottleneck Feature-Based Hybrid Deep Autoencoder Approach for Indian Language Identification
【24h】

Bottleneck Feature-Based Hybrid Deep Autoencoder Approach for Indian Language Identification

机译：基于瓶颈特征的混合深度自动编码器用于印度语言识别

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Latest and emerging approaches are essential to resolve the communication barrier among different languages in speechprocessing. The automatic language identification system is developed to identify the spoken language from speech utterances.Feature selection is a very challenging task in language identification. In this paper, bottleneck feature-based hybrid deepautoencoder approach is proposed to identify the given speech signal with corresponding language features. In the proposedapproach, initially Mel-frequency cepstral coefficients, linear prediction coefficients, and shifted delta coefficients featuresare directly extracted from multilingual speech utterances. Further, we extracted bottleneck feature from the bottleneck layerof the bottleneck deep neural network. Initially, recognition rate has been evaluated for each feature set to find out the bestfeature. Finally, the best feature along with other features is used as the input for deep autoencoder with softmax regressionto identify the language based on class labels. The deep autoencoder is fine-tuned to reach the global optimum through Jayaoptimization algorithm. To carry out the experiments, the recorded database is used for four Indian languages with specialemphasis on northeastern languages. The experimental results demonstrate that the proposed hybrid approach using bottleneckfeature with shifted delta coefficients is performing well with 97.10% accuracy. Moreover, experimental results also show thatproposed hybrid approach gives superior outcome when compared with the baseline deep neural network-based approaches.
机译：最新的和新兴的方法对于解决语音处理中不同语言之间的交流障碍至关重要。开发自动语言识别系统以从语音中识别口语。特征选择是语言识别中一项非常艰巨的任务。本文提出了一种基于瓶颈特征的混合深度自动编码器方法，以识别具有相应语言特征的语音信号。在提出的方法中，首先从多语言语音中直接提取梅尔频率倒谱系数，线性预测系数和位移增量系数特征。此外，我们从瓶颈深度神经网络的瓶颈层提取了瓶颈特征。最初，已经对每个功能集的识别率进行了评估，以找出最佳功能。最后，最佳功能与其他功能一起用作具有softmax回归的深度自动编码器的输入，以基于类标签识别语言。深度自动编码器通过Jayaoptimization算法进行了微调，以达到全局最优。为了进行实验，记录的数据库用于四种印度语言，其中东北语言特别强调。实验结果表明，所提出的使用瓶颈特征和偏移增量系数的混合方法以97.10％的精度表现良好。此外，实验结果还表明，与基于基线深层神经网络的方法相比，提出的混合方法可提供更好的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号