復(fù)雜背景下聲紋特征提取與識(shí)別
發(fā)布時(shí)間：2018-11-02 09:31
【摘要】：隨著互聯(lián)網(wǎng)以及信息化的迅速發(fā)展,聲紋識(shí)別技術(shù)在金融、證券、社保、電子商務(wù)、銀行等遠(yuǎn)程客戶服務(wù)的身份確認(rèn)和公安、軍隊(duì)安全部門的特定人身份自動(dòng)檢測(cè)和認(rèn)證中具有廣泛的應(yīng)用價(jià)值和前景需求,是當(dāng)今世界聲音信號(hào)處理和生物特征信息檢測(cè)與識(shí)別領(lǐng)域的重要探索方向。近幾十年來,在這一領(lǐng)域的研究已經(jīng)取得了重大進(jìn)展,但因?yàn)檎f話人個(gè)性特征易受外界因素影響以及具體實(shí)際環(huán)境的復(fù)雜多變性,其瓶頸效應(yīng)也逐漸凸顯,因此,在復(fù)雜背景下研究有效的語音信息檢測(cè)方法和更具魯棒性的特征提取算法對(duì)于提高系統(tǒng)的識(shí)別率具有非常重要的意義。 復(fù)雜背景下的聲紋識(shí)別技術(shù)是在高度復(fù)雜噪聲情況下,通過檢測(cè)出聲音并進(jìn)一步進(jìn)行特征提取后,經(jīng)過分析處理建立識(shí)別模型,最后應(yīng)用識(shí)別模型對(duì)說話人進(jìn)行識(shí)別。論文主要研究語音端點(diǎn)檢測(cè)方法和特征提取方法來提高識(shí)別效率,主要工作如下。 首先,在聲音預(yù)處理階段,提出了嘈雜環(huán)境下的兩種語音信號(hào)端點(diǎn)檢測(cè)方法,根據(jù)不同背景復(fù)雜程度的信噪比高低分別采用基于譜熵的端點(diǎn)檢測(cè)算法和基于短時(shí)能量和過零率的雙門限端點(diǎn)檢測(cè)算法,實(shí)驗(yàn)表明,背景為高信噪比情況下基于短時(shí)能量和過零率的雙門限端點(diǎn)檢測(cè)算法效果較好,背景為低信噪比情況下基于譜熵的端點(diǎn)檢測(cè)算法較優(yōu)。 其次,在特征提取階段,利用倒譜法計(jì)算出基音周期參數(shù),再通過Mel濾波器組將語音信號(hào)功率譜轉(zhuǎn)換成Mel倒譜系數(shù)(MFCC),然后利用改進(jìn)特征提取算法將兩種參數(shù)組成一種聲紋特征參量,同時(shí)分別對(duì)它們進(jìn)行了實(shí)驗(yàn)仿真。 最后,在聲紋識(shí)別階段,首先提出帶噪特征的識(shí)別算法(SEMG)算法,即在復(fù)雜背景下對(duì)語音信號(hào)利用基于譜熵的端點(diǎn)檢測(cè)算法檢測(cè)端點(diǎn)后,再利用改進(jìn)特征提取算法特征提取,最后為每個(gè)說話人建立一個(gè)高斯混合模型(GMM),并通過實(shí)驗(yàn)驗(yàn)證了SEMG算法的有效性,達(dá)到了理想結(jié)果。
[Abstract]:With the rapid development of the Internet and information technology, voiceprint identification technology in finance, securities, social security, e-commerce, banking and other remote customer service identification and public security, The automatic detection and authentication of the specific identity of the military security department has a wide range of application value and foreground requirements. It is an important exploration direction in the field of sound signal processing and biometric information detection and recognition in the world today. In recent decades, great progress has been made in the research in this field. However, because the speaker's personality is easily influenced by the external factors and the complex variability of the actual environment, the bottleneck effect is becoming more and more prominent. It is very important to study the effective speech information detection method and the more robust feature extraction algorithm in complex background for improving the recognition rate of the system. The voiceprint recognition technology in complex background is based on the detection of sound and further feature extraction. After analyzing and processing, the recognition model is established. Finally, the recognition model is used to recognize the speaker. This paper mainly studies the speech endpoint detection method and feature extraction method to improve the recognition efficiency, the main work is as follows. Firstly, in the stage of sound preprocessing, two speech signal endpoint detection methods in noisy environment are proposed. According to the signal-to-noise ratio of different background complexity, the two threshold endpoint detection algorithms based on spectral entropy and short-time energy and zero-crossing rate are used, respectively. The experimental results show that, The dual-threshold endpoint detection algorithm based on short-time energy and zero-crossing rate is better in the case of high signal-to-noise ratio (SNR), and the algorithm based on spectral entropy is better when the background is low SNR. Secondly, in feature extraction stage, pitch period parameters are calculated by cepstrum method, and then the power spectrum of speech signal is converted to Mel cepstrum coefficient (MFCC), by Mel filter bank. Then, the improved feature extraction algorithm is used to make two parameters into one voiceprint feature parameter, and at the same time, the experimental simulation of them is carried out. Finally, in the stage of voiceprint recognition, a noisy feature recognition algorithm (SEMG) is proposed, that is, the speech signal is detected by spectral entropy based endpoint detection algorithm under complex background, and then the improved feature extraction algorithm is used to extract features. Finally, a Gao Si hybrid model, (GMM), is established for each speaker, and the effectiveness of the SEMG algorithm is verified by experiments, and the ideal results are obtained.
【學(xué)位授予單位】：中南林業(yè)科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TN912.34
本文編號(hào)：2305639
[Abstract]:With the rapid development of the Internet and information technology, voiceprint identification technology in finance, securities, social security, e-commerce, banking and other remote customer service identification and public security, The automatic detection and authentication of the specific identity of the military security department has a wide range of application value and foreground requirements. It is an important exploration direction in the field of sound signal processing and biometric information detection and recognition in the world today. In recent decades, great progress has been made in the research in this field. However, because the speaker's personality is easily influenced by the external factors and the complex variability of the actual environment, the bottleneck effect is becoming more and more prominent. It is very important to study the effective speech information detection method and the more robust feature extraction algorithm in complex background for improving the recognition rate of the system. The voiceprint recognition technology in complex background is based on the detection of sound and further feature extraction. After analyzing and processing, the recognition model is established. Finally, the recognition model is used to recognize the speaker. This paper mainly studies the speech endpoint detection method and feature extraction method to improve the recognition efficiency, the main work is as follows. Firstly, in the stage of sound preprocessing, two speech signal endpoint detection methods in noisy environment are proposed. According to the signal-to-noise ratio of different background complexity, the two threshold endpoint detection algorithms based on spectral entropy and short-time energy and zero-crossing rate are used, respectively. The experimental results show that, The dual-threshold endpoint detection algorithm based on short-time energy and zero-crossing rate is better in the case of high signal-to-noise ratio (SNR), and the algorithm based on spectral entropy is better when the background is low SNR. Secondly, in feature extraction stage, pitch period parameters are calculated by cepstrum method, and then the power spectrum of speech signal is converted to Mel cepstrum coefficient (MFCC), by Mel filter bank. Then, the improved feature extraction algorithm is used to make two parameters into one voiceprint feature parameter, and at the same time, the experimental simulation of them is carried out. Finally, in the stage of voiceprint recognition, a noisy feature recognition algorithm (SEMG) is proposed, that is, the speech signal is detected by spectral entropy based endpoint detection algorithm under complex background, and then the improved feature extraction algorithm is used to extract features. Finally, a Gao Si hybrid model, (GMM), is established for each speaker, and the effectiveness of the SEMG algorithm is verified by experiments, and the ideal results are obtained.
【學(xué)位授予單位】：中南林業(yè)科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TN912.34
【參考文獻(xiàn)】
相關(guān)期刊論文 前9條
1 鄧浩江,王守覺,邢藏菊,李倩;基于聚類統(tǒng)計(jì)與文本無關(guān)的說話人識(shí)別研究[J];電路與系統(tǒng)學(xué)報(bào);2001年03期
2 賀永強(qiáng);石艷榮;;用CA認(rèn)證技術(shù)解決遠(yuǎn)程教育中的身份認(rèn)證[J];計(jì)算機(jī)安全;2009年05期
3 李虎生,劉加,劉潤生;語音識(shí)別說話人自適應(yīng)研究現(xiàn)狀及發(fā)展趨勢(shì)[J];電子學(xué)報(bào);2003年01期
4 張文耀,許剛,王裕國;循環(huán)AMDF及其語音基音周期估計(jì)算法[J];電子學(xué)報(bào);2003年06期
5 白瑩;趙振東;戚銀城;王斌;郭建勇;;基于小波神經(jīng)網(wǎng)絡(luò)的與文本無關(guān)說話人識(shí)別方法研究[J];電子與信息學(xué)報(bào);2006年06期
6 孫妍;楊曉非;;基于MFCC和LSP混合的語音特征參數(shù)的技術(shù)研究[J];計(jì)算機(jī)與信息技術(shù);2007年Z1期
7 盧官明;李海波;劉莉;;生物特征識(shí)別綜述[J];南京郵電大學(xué)學(xué)報(bào)(自然科學(xué)版);2007年01期
8 趙振東;張靜;李圓;胡喜梅;;基于GMM說話人分類的說話人識(shí)別方法研究[J];通信技術(shù);2009年10期
9 王偉;鄧輝文;;基于MFCC參數(shù)和VQ的說話人識(shí)別系統(tǒng)[J];儀器儀表學(xué)報(bào);2006年S3期
，本文編號(hào)：2305639
本文鏈接：http://sikaile.net/kejilunwen/wltx/2305639.html
最近更新
教材專著
