面向真实噪声环境的语种识别

doi:10.13190/j.jbupt.2021-053

北京邮电大学学报 ›› 2021, Vol. 44 ›› Issue (6): 134-140.doi: 10.13190/j.jbupt.2021-053

• 研究报告 • 上一篇

面向真实噪声环境的语种识别

邵玉斌, 刘晶, 龙华, 李一民

昆明理工大学信息工程与自动化学院, 昆明 650500

收稿日期:2021-04-02 出版日期:2021-12-28 发布日期:2021-12-28
通讯作者: 刘晶(1996—),男,硕士生,E-mail:liujing@stu.kust.edu.cn. E-mail:liujing@stu.kust.edu.cn
作者简介:邵玉斌(1970—),男,教授,硕士生导师.
基金资助:
国家自然科学基金项目(61761025)

Language Identification in Real Noisy Environments

SHAO Yu-bin, LIU Jing, LONG Hua, LI Yi-min

Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China

Received:2021-04-02 Online:2021-12-28 Published:2021-12-28

摘要/Abstract

摘要： 语种识别受真实噪声环境的影响较大,识别效果不佳. 为了解决真实噪声环境下语种识别的问题,提出一种基于对数灰度语谱图的图像处理语种识别方法. 根据噪声能量和语音能量在语谱图上的分布规律对真实噪声中的语音信号进行带通滤波;再结合人耳听觉特性提取对数灰度语谱图;然后提取图像主成分特征作为语种特征,采用残差神经网络模型进行训练测试. 实验结果表明,在掠夺者战斗机驾驶舱的环境下,所提方法的平均识别正确率相对于线性灰度语谱图方法提升了27.5%,在其他噪声环境下的平均识别正确率也有提升.

关键词: 语种识别, 真实噪声环境, 对数灰度语谱图, 残差神经网络, 图像处理

Abstract: Language identification is heavily influenced by the real noise environment, resulting in poor identification results. To solve this problem, an image processing method for language identification is proposed based on the logarithmic gray-scale speech spectrogram. The logarithmic gray-scale speech spectrogram is obtained by combining the human auditory properties and the voice filtered in real noise environments according to the different distribution of noise and speech on the speech spectrogram. Then, the principal component of the spectrogram is extracted as language features and, a residual neural network model is used for training and testing. Experimental results show that the average identification rate of the proposed method is improved by 27.5% in the noisy cockpit of a Blackburn Buccaneer compared to the linear grey-scale speech spectrogram method. In other noisy environments, the average identification rate is also improved.

Key words: language identification, real noise environment, logarithmic gray-scale speech spectrogram, residual neural network, image processing

中图分类号:

TN912.3

邵玉斌, 刘晶, 龙华, 李一民. 面向真实噪声环境的语种识别[J]. 北京邮电大学学报, 2021, 44(6): 134-140.

SHAO Yu-bin, LIU Jing, LONG Hua, LI Yi-min. Language Identification in Real Noisy Environments[J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(6): 134-140.

参考文献

[1] Li Haizhou, Ma Bin, Lee K A.Spoken language recognition:from fundamentals to practice[J].Proceedings of the IEEE, 2013, 101(5):1136-1159.
[2] 张卫强, 刘加.基于听感知特征的语种识别[J].清华大学学报(自然科学版), 2009, 49(1):78-81.Zhang Weiqiang, Liu Jia.Language recognition based on auditory perception characteristics[J].Journal of Tsinghua University (Natural Science Edition), 2009, 49(1):78-81.
[3] Zissman M A.Comparison of four approaches to automa-tic language identification of telephone speech[J].IEEE Transactions on Speech and Audio Processing, 1996, 4(1):31.
[4] Montavon G.Deep learning for spoken language identification[C]//2009 NIPS Workshop on Deep Learning for Speech Recognition and Related Applications.Vancouver:NIPS Foundation, 2009:1-4.
[5] Jiang Bing, Song Yan, Wei Si, et al.Performance evaluation of deep bottleneck features for spoken language identification[C]//The 9th International Symposium on Chinese Spoken Language Processing.Singapore:IEEE, 2014:143-147.
[6] Lopez-Moreno I, Gonzalez-Dominguez J, Plchot O, et al.Automatic language identification using deep neural networks[C]//2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).Florence:IEEE Press, 2014:5337-5341.
[7] Geng Wang, Wang Wenfu, Zhao Yuanyuan, et al.End-to-end language identification using attention-based recurrent neural networks[C]//Interspeech 2016.San Francisco:ISCA, 2016:2944-2948.
[8] Jin Ma, Song Yan, Mcloughlin I, et al.LID-senones and their statistics for language identification[J].IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2018, 26(1):171-183.
[9] Cai Weicheng, Cai Zexin, Liu Wenbo, et al.Insights into end-to-end learning scheme for language identification[J].IEEE Signal Processing Society Sigport, 2018, 28(2):202-210.
[10] Deshwal D, Sangwan P, Kumar D.Feature extraction methods in language identification:a survey[J].Wireless Personal Communications, 2019, 107(4):2071-2103.
[11] Li Lin, Li Zheng, Liu Yan, et al.Deep joint learning for language recognition[J].Neural Networks:the Official Journal of the International Neural Network Society, 2021, 141(9):72-86.
[12] Bhanja C C, Laskar M A, Laskar R H.Modelling multi-level prosody and spectral features using deep neural network for an automatic tonal and non-tonal pre-classification-based Indian language identification system[J].Language Resources and Evaluation, 2021, 55(3):689-730.
[13] He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE Press, 2016:770-778.
[14] 刘威.单通道语音水印与语音增强算法研究[D].南京:东南大学, 2017.
[15] Franzoni V, Biondi G, Milani A.Emotional sounds of crowds:spectrogram-based analysis using deep learning[J].Multimedia Tools and Applications, 2020, 79(47/48):36063-36075.
[16] 蓝雯飞, 汪敦志, 张盛兰.一种新的降维算法PCA_LLE在图像识别中的应用[J].中南民族大学学报(自然科学版), 2020, 39(1):85-90.Lan Wenfei, Wang Dunzhi, Zhang Shenglan.Application of a new dimensionality reduction algorithm PCA_LLE in image recognition[J].Journal of South-Central University for Nationalities (Natural Science Edition), 2020, 39(1):85-90.
[17] Qaraei M, Abbaasi S, Ghiasi-Shirazi K.Randomized non-linear PCA networks-science direct[J].Information Sciences, 2021, 545:241-253.
[18] Zhu Dong, Huang Ming, Yang Jingjing, et al.Identification of spoken language from webcast using deep con-volutional recurrent neural networks[C]//2019 International Conference on Information Technology.Sanya:Electrical and Electronic Engineering (ITEEE 2019), 2019:1147-1152.

面向真实噪声环境的语种识别

Language Identification in Real Noisy Environments

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价

[1]	王延凯龙华邵玉斌杜庆治王瑶. 基于融合特征MGCC的语种识别方法[J]. 北京邮电大学学报, 2023, 46(2): 116-121.
[2]	张昊阁邵玉斌龙华杜庆治周大春. 基于Gammatone尺度功率规整系数谱图的语种识别[J]. 北京邮电大学学报, 2023, 46(2): 122-128.
[3]	张昊阁邵玉斌龙华彭艺周大春. 基于对数Gammatone滤波器能量谱图的语种识别[J]. 北京邮电大学学报, 2023, 46(1): 38-43.
[4]	邵玉斌, 刘晶, 龙华, 杜庆治, 李一民. 基于声道频谱参数的语种识别[J]. 北京邮电大学学报, 2021, 44(3): 112-119.
[5]	逄增治, 史建杰, 尹建芹, 朱利民, 李金屏. 基于目标形态特征的工件自动分割方法[J]. 北京邮电大学学报, 2019, 42(5): 119-126.
[6]	赵珊, 李永思. 基于主成分分析的图像哈希算法[J]. 北京邮电大学学报, 2019, 42(2): 36-41.
[7]	周东, 李昀峰, 高妍, 阮雅端, 陈启美. 融合区域纹理梯度的船舶阴影视频去除算法[J]. 北京邮电大学学报, 2017, 40(s1): 81-84,97.
[8]	绽琨张红娟马义德刘丽田乐. 交叉皮层模型及其在图像处理中的应用[J]. 北京邮电大学学报, 2009, 32(4): 40-45.
[9]	王建勇,周晓光,廖启征. 基于2级检测的脉冲噪声滤除算法[J]. 北京邮电大学学报, 2005, 28(3): 59-61.
[10]	赵艳明, 全子一. 一种有效的小波-Wiener滤波去噪算法[J]. 北京邮电大学学报, 2004, 27(4): 41-45.
[11]	邓国强, 孙景鳌, 蔡安妮, 董守平. 一种基于曲线积分的区域填充算法[J]. 北京邮电大学学报, 2001, 24(2): 87-91.
[12]	施可为, 蔡安妮, 孙景鳌. H.323视频会议终端的软件实现[J]. 北京邮电大学学报, 2000, 23(3): 35-38.
[13]	林国辉,门爱东,全子一. 一种基于块边界连续规则的后处理算法[J]. 北京邮电大学学报, 2000, 23(2): 87-90.