The unequal error protection （UEP） is applied in distributed speech recognition （DSR） system and three schemes are proposed. All of these three schemes are evaluated on the GSM simulating platform for recognizing ...The unequal error protection （UEP） is applied in distributed speech recognition （DSR） system and three schemes are proposed. All of these three schemes are evaluated on the GSM simulating platform for recognizing mandarin digit strings and compared with the equal error protection （EEP） scheme. Experiments show that UEP can protect the data transmitted in DSR system more effectively, which results in a higher word accurate rate of DSR system.展开更多
Data-driven temporal filtering technique is integrated into the time trajectory of Teager energy operation （TEO） based feature parameter for improving the robustness of speech recognition system against noise. Three...Data-driven temporal filtering technique is integrated into the time trajectory of Teager energy operation （TEO） based feature parameter for improving the robustness of speech recognition system against noise. Three kinds of data-driven temporal filters are investigated for the motivation of alleviating the harmful effects that the environmental factors have on the speech. The filters include： principle component analysis （PCA） based filters, linear discriminant analysis （LDA） based filters and minimum classification error （MCE） based filters. Detailed comparative analysis among these temporal filtering approaches applied in Teager energy domain is presented. It is shown that while all of them can improve the recognition performance of the original TEO based feature parameter in adverse environment, MCE based temporal filtering can provide the lowest error rate as SNR decreases than any other algorithms.展开更多
With the increasing requirement of military and security, the technology of information hiding for speech becomes a hotspot and difficulty in the fields of speech signal processing and in-formation security, which is ...With the increasing requirement of military and security, the technology of information hiding for speech becomes a hotspot and difficulty in the fields of speech signal processing and in-formation security, which is developing rapidly. In order to stand against the stegano-analysis, the paper proposed an optimal information hiding algorithm for speech in the Fractional Fourier Transform (FrFT) domain based on the Minimum Mean Square Error (MMSE) criterion. The results of simulation and experiments show that speech modified by the proposed algorithm has no remarkable changes both in time and frequency domains, which can effectively resist the time and frequency analysis, Otherwise, the algorithm is robust to general signal process attack, and the difference is imperceptible between the original and modified speech.展开更多
An improved method based on minimum mean square error-short time spectral amplitude （MMSE-STSA） is proposed to cancel background noise in whispered speech. Using the acoustic character of whispered speech, the algor...An improved method based on minimum mean square error-short time spectral amplitude （MMSE-STSA） is proposed to cancel background noise in whispered speech. Using the acoustic character of whispered speech, the algorithm can track the change of non-stationary background noise effectively. Compared with original MMSE-STSA algorithm and method in selectable mode Vo-coder （SMV）, the improved algorithm can further suppress the residual noise for low signal-to-noise radio （SNR） and avoid the excessive suppression. Simulations show that under the non-stationary noisy environment, the proposed algorithm can not only get a better performance in enhancement, but also reduce the speech distortion.展开更多
传统的跨语种交互翻译机器人语义纠错方法通常是单向的,效率较低,导致识别错误率较高。为此,文章提出基于语音信号的跨语种交互翻译机器人语义纠错方法。在基础语音识别的基础上,通过交互标定和特征提取来修正语义错误位置,并设计语音...传统的跨语种交互翻译机器人语义纠错方法通常是单向的,效率较低,导致识别错误率较高。为此,文章提出基于语音信号的跨语种交互翻译机器人语义纠错方法。在基础语音识别的基础上,通过交互标定和特征提取来修正语义错误位置,并设计语音信号翻译机器人的语义纠错模型,采用随时间反向传播(Backpropagation Through Time,BPTT)循环训练核验方式,以确保纠错的准确性。测试结果显示,经过3个阶段测试,选定的5段语音材料的纠错识别率成功控制在10%以下,表明基于语音信号的跨语种交互翻译机器人语义纠错方法高效,具有实际应用价值。展开更多
基金Sponsored bythe National Natural Science Foundation of China (60372089) the Basic Research Foundation of Beijing Institute of Technology(BIT-UBF-200301F03)
文摘The unequal error protection （UEP） is applied in distributed speech recognition （DSR） system and three schemes are proposed. All of these three schemes are evaluated on the GSM simulating platform for recognizing mandarin digit strings and compared with the equal error protection （EEP） scheme. Experiments show that UEP can protect the data transmitted in DSR system more effectively, which results in a higher word accurate rate of DSR system.
基金Supported by National High Technology Research and Development Program of China （863 Program） （2008AA040201）, National Natural Science Foundation of China （90920302）, National Science and Technology Pillar Program of China （2009BAH41B01）, National Natural Science Foundation of China and Research Grants Council of Hong Kong （60931160443） The authors thank Michael T. Johnson in the Depart- ment of Electrical Engineering, Marquette University in USA for the experiments suggestion and helping to improve the English writing.
基金Sponsored bythe Basic Research Foundation of Beijing Institute of Technology (BIT-UBF-200301F03) BIT &Ericsson Cooperation Project
文摘Data-driven temporal filtering technique is integrated into the time trajectory of Teager energy operation （TEO） based feature parameter for improving the robustness of speech recognition system against noise. Three kinds of data-driven temporal filters are investigated for the motivation of alleviating the harmful effects that the environmental factors have on the speech. The filters include： principle component analysis （PCA） based filters, linear discriminant analysis （LDA） based filters and minimum classification error （MCE） based filters. Detailed comparative analysis among these temporal filtering approaches applied in Teager energy domain is presented. It is shown that while all of them can improve the recognition performance of the original TEO based feature parameter in adverse environment, MCE based temporal filtering can provide the lowest error rate as SNR decreases than any other algorithms.
基金Supported by the National Natural Science Foundation of China (No. 60472058, No. 60975017)Jiangsu Provincial Natural Science Foundation (No. BK2008291)
文摘With the increasing requirement of military and security, the technology of information hiding for speech becomes a hotspot and difficulty in the fields of speech signal processing and in-formation security, which is developing rapidly. In order to stand against the stegano-analysis, the paper proposed an optimal information hiding algorithm for speech in the Fractional Fourier Transform (FrFT) domain based on the Minimum Mean Square Error (MMSE) criterion. The results of simulation and experiments show that speech modified by the proposed algorithm has no remarkable changes both in time and frequency domains, which can effectively resist the time and frequency analysis, Otherwise, the algorithm is robust to general signal process attack, and the difference is imperceptible between the original and modified speech.
文摘An improved method based on minimum mean square error-short time spectral amplitude （MMSE-STSA） is proposed to cancel background noise in whispered speech. Using the acoustic character of whispered speech, the algorithm can track the change of non-stationary background noise effectively. Compared with original MMSE-STSA algorithm and method in selectable mode Vo-coder （SMV）, the improved algorithm can further suppress the residual noise for low signal-to-noise radio （SNR） and avoid the excessive suppression. Simulations show that under the non-stationary noisy environment, the proposed algorithm can not only get a better performance in enhancement, but also reduce the speech distortion.
文摘传统的跨语种交互翻译机器人语义纠错方法通常是单向的,效率较低,导致识别错误率较高。为此,文章提出基于语音信号的跨语种交互翻译机器人语义纠错方法。在基础语音识别的基础上,通过交互标定和特征提取来修正语义错误位置,并设计语音信号翻译机器人的语义纠错模型,采用随时间反向传播(Backpropagation Through Time,BPTT)循环训练核验方式,以确保纠错的准确性。测试结果显示,经过3个阶段测试,选定的5段语音材料的纠错识别率成功控制在10%以下,表明基于语音信号的跨语种交互翻译机器人语义纠错方法高效,具有实际应用价值。
