基于数字对讲机的MELP声码器算法研究与实现
摘要
在通信系统全面数字化的趋势下,模拟对讲机也加速了向数字对讲机过渡的进程。专网通信领域内数字对讲机正以其强劲的优势逐步取代模拟对讲机。各个国际领先的数字对讲厂商也加入了数字对讲机的研究,共同推进数字对讲机的发展。而与此同时,语音作为任何一个通信系统中的主导业务之一,语音编码技术也成为了各个通信系统的核心技术之一。在数字对讲机中,声码器算法的设计和研究也是其核心部分。本文基于DPMR和DMR标准,将MELP声码器算法应用于数字对讲机中,对MELP声码器的设计与实现进行了研究。
首先,分析了数字对讲机的发展状况,对数字对讲机的主流标准协议进行了简单的介绍。同时对语音编码进行了基本的介绍。
其次,对MELP声码器算法的分析基础进行了详细的讨论,在语音数字模型的基础上,详细研究了线性预测分析。同时研究了本算法中的LP参数的求解方法,并详细推导了求LSF参数的转化算法。
最后,完成了MELP声码器算法的设计与实现。对MELP算法进行了详尽的分析研究,建立MATLAB算法仿真模型分析示例语音。用标准C语言程序编程实现本算法,且用主观语音质量PESQ测试合成语音的客观MOS得分,并给出了优化算法的方法。最后在算法的基础上设计了一种基于DMR标准协议的改进算法,通过验证,改进算法方案可行。
As the overall digital transformation in communication systems, analog radio also accelerated the transitional process to digital radio. In the field of private network communication, digital radio is gradually replacing analog radio with its strong advantages. Various international manufacturers in digital intercom have joined the digital radio research, to jointly promote the development of digital radio. Meanwhile, the voice operation as one of the leading business in any communication system, speech coding technology has also become the core technology of each one communication system. In digital radio, the research and design of voice coder algorithm is also the core portion. In this paper, based on DPMR (Digital Private Mobile Radio) and DMR (Digital Mobile Radio) standard, the MELP (Mixed Excitation Linear Prediction) vocoder algorithm is applied to digital radio. And also we have studied the research and realization of MELP vocoder.
Firstly, the paper analyzes the development of digital radio, while briefly introduces the mainstream standard protocol of the digital radio. At the same time, conducts a basic introduction for speech coding.
Secondly, this article detailedly discusses the analysis basis of MELP vocoder algorithm. Based on the speech digital model, we have studied the LP (linear prediction) analysis. The paper proposes the method for solving the LP parameters in the algorithm, and derives the transformation algorithm for seeking LSF (Line spectrum Frequency) parameters.
Finally, the paper presents the design and implementation of MELP vocoder algorithm. We have thoroughly analyzed the MELP algorithm, and obtained the parameters of analysis voice by building the MATLAB simulaton model. And also we use the standard C language program achieve this algorithm. Then test the synthesized speech with the PESQ (Perceptual evaluation of speech quality) and get the objective MOS score. And the method of optimizing the algorithm is given. At the end, on the basis of the algorithm an improved algorithm is presented which is based on DMR standard protocol, and by verifying, the improved algorithm is feasible.
首先,分析了数字对讲机的发展状况,对数字对讲机的主流标准协议进行了简单的介绍。同时对语音编码进行了基本的介绍。
其次,对MELP声码器算法的分析基础进行了详细的讨论,在语音数字模型的基础上,详细研究了线性预测分析。同时研究了本算法中的LP参数的求解方法,并详细推导了求LSF参数的转化算法。
最后,完成了MELP声码器算法的设计与实现。对MELP算法进行了详尽的分析研究,建立MATLAB算法仿真模型分析示例语音。用标准C语言程序编程实现本算法,且用主观语音质量PESQ测试合成语音的客观MOS得分,并给出了优化算法的方法。最后在算法的基础上设计了一种基于DMR标准协议的改进算法,通过验证,改进算法方案可行。
As the overall digital transformation in communication systems, analog radio also accelerated the transitional process to digital radio. In the field of private network communication, digital radio is gradually replacing analog radio with its strong advantages. Various international manufacturers in digital intercom have joined the digital radio research, to jointly promote the development of digital radio. Meanwhile, the voice operation as one of the leading business in any communication system, speech coding technology has also become the core technology of each one communication system. In digital radio, the research and design of voice coder algorithm is also the core portion. In this paper, based on DPMR (Digital Private Mobile Radio) and DMR (Digital Mobile Radio) standard, the MELP (Mixed Excitation Linear Prediction) vocoder algorithm is applied to digital radio. And also we have studied the research and realization of MELP vocoder.
Firstly, the paper analyzes the development of digital radio, while briefly introduces the mainstream standard protocol of the digital radio. At the same time, conducts a basic introduction for speech coding.
Secondly, this article detailedly discusses the analysis basis of MELP vocoder algorithm. Based on the speech digital model, we have studied the LP (linear prediction) analysis. The paper proposes the method for solving the LP parameters in the algorithm, and derives the transformation algorithm for seeking LSF (Line spectrum Frequency) parameters.
Finally, the paper presents the design and implementation of MELP vocoder algorithm. We have thoroughly analyzed the MELP algorithm, and obtained the parameters of analysis voice by building the MATLAB simulaton model. And also we use the standard C language program achieve this algorithm. Then test the synthesized speech with the PESQ (Perceptual evaluation of speech quality) and get the objective MOS score. And the method of optimizing the algorithm is given. At the end, on the basis of the algorithm an improved algorithm is presented which is based on DMR standard protocol, and by verifying, the improved algorithm is feasible.
引文
[1]百度百科.第三代移动通信[EB/OL]. http://baike.baidu.com/view/138792.htm#sub138792
[2]王洪,唐凯.低速率语音编码[M].北京：国防工业出版社,2005：122-148
[3]戚银城,张巍,苑津莎.超低速率MELP语音编码算法研究[J].声学技术,2007,26(6)：1196-1200
[4]雒国成,蔡芳.基于MELP算法的语音编解码器分析[J],黄河水利职业技术学院学报,2008,20(1)：49-51
[5]张毓忠,梅武钢.语音信号MELP编码的研究与实现[J].电声技术,2000(11)：6-9
[6]马震,蒋保臣,陶立慧.混合激励线性预测(MELP)方法研究与改进[J].现代电子技术,2005(13)：22-25
[7]陈双燕,张铁军,王东辉,侯朝焕.基于一种可配置可扩展处理器的MELP语音算法的改进与实现[J].微电子学与计算机,2006,23(6)：42-44
[8]王炳锡.语音编码[M].西安：西安电子科技人学出版社,2002：257-274
[9]徐贵森,刘鑫,潭学治.浅谈无线对讲机模拟转数字[J].移动通信,2007：22-25
[10]李慧芳.基于MELP的低码率语音编码技术研究与实现[D].北京：北京邮电大学,2008
[11]高浩.基于DSP关于改进的MBE语音算法的研究[D].江苏：江南大学,2006年
[12]王宇寰.基于声码器的900MHz数字对讲机研究[D].西安：西安电子科技大学,2007
[13]李娜,王中元,胡瑞敏,朱福喜.数字语音编码技术和标准介绍[J].中国新通信(技术版),2007(13)：3943
[14]程飞.基于MELP的高效声码器研究[D].北京：国防科学技术大学,2006
[15]史水平,李世作.线性预测编码技术及其在音频文件上的应用[J].现代电子技术,2004(4)：21-23
[16]洪治.我国对讲机市场现状及发展趋势[J].中国无线电,2007(1)：37,64
[17]刘雁飞.数字无中心多信道选址移动通信协议的研究[D].北京：北京化工大学,2008
[18]ETSITS 102 361-1. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR)Systems; Part1:DMR Air Interface (AI) protocol"[S].2007
[19]ETSI TS 102 361-2. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part2:DMR voice and generic services and facilities" [S].2007
[20]ETSI TS 102 361-3. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part 3:DMR Data protocol"[S].2007
[21]ETSI TS 102 361-4. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part 4:DMR trunking protocol"[S].2007
[22]李进良.对制订我国数字专业移动通信标准的建议[J].移动通信,2008,17：9-13
[23]王保坤.DMR端机基带模块研究与实现[D].西安：西安电子科技大学,2010
[24]百度百科.多址方式[EB/OL]. http://baike.baidu.com/view/184175.htm
[25]李志芳.数字对讲机——对讲机技术发展的必然方向[J].现代通信,2009
[26]刘辉.DMR数字对讲机协议的研究与实现[D].北京：北京化工大学,2010
[27]ETSI TS 300 168. "Radio Equipment and Systems (RES); Digital Short Range Radio (DSRR)" [S].1993
[28]Alan V. McCree, Thomas P. Barnwell Ⅲ. A Mixed Excitation LPC Vocoder Model for Low Bit Rate Speech Coding[J]. IEEE Transaction On speech and audio processing,1995,3(4):242-250
[29]姚天任.数字语音处理[M].武汉：华中科技大学出版社,2007
[30]张旭.多带激励(MBE)低速率语音编码技术的研究与实现[D].天津：天津大学,2005
[31]易克初.语音信号处理[M].北京：国防工业出版社,2000：15-115
[32]刘媛媛.基于DSP的MELP语音编码算法研究及实现[D].哈尔滨：哈尔滨工程大学,2008
[33]杨行峻,迟惠生.语音信号数字处理[M].北京：电子工业出版社,1995：48-127
[34]Makhoul J. I. Linear Prediction:A Tutorial Review[J]. Proc. IEEE,1995,63(4):561-580.
[35]张毓忠,梅武钢.语音信号MELP编码的研究与实现[J].电声技术,2000(11)：6-9
[36]马震,蒋保臣,陶立慧.混合激励线性预测(MELP)方法研究与改进[J].现代电子技术,2005(13)：22-25
[37]戚银城,张巍,苑津莎.超低速率MELP语音编码算法研究[J].声学技术,2007(26)：1196-1200
[38]刘宽海,葛万成.MELP低比特率数字语音编码技术研究[J].信息技术,2003,27(8)：52-54
[39]张雄伟.2.4kb/s MELP低速率语音编码算法中的关键技术[J].电子工程师,1999(4)：21-23
[40 T. Tremain. The government standard linear Predictive coding algorithm:LPC-10[J]. Speech Technology Magazine,1982:40-49
[41]冷丰麟.基于声码化技术的数字语音加密系统的研究[D].大连：大连海事大学,2009
[42]D. W. Griffin. Multi-band excitation vocoder[J]. IEEE Transaction on ASSP,1988,36(8): 1223-1235
[43]D. W. Griffin, J. S. Lim. Signal estimation from modified short-time fourier transform[J]. IEEE Transaction On ASSP,1984,32(2):236-243
[44]P. C. Meuse. A 2400 bps multi-band excitation speech coder[J]. ICASSP,1990(88):9-12
[45]Griffith, D.W., Lim, J.S.. Multiband excitation vocoder[J]. IEEE Transactions on ASSP,1988, 36(8):1223-1235
[46]Alan V. McCree, E. Bryan George, Thomas P. Barnwell. A 2.4 kbit/s MELP coder candidate for the new U. S. Federal Standard[J]. Proceeding for ICASSP Atlanta,1996,1(1):200-203
[47]王都生.多带混合激励低速率语音编码的研究[D].西安：西安电子科技大学,1998
[48]胡剑凌,徐盛,陈健.2.4kb/sMELP算法设计[J].上海交通大学学报,2000,34(6)：789-792
[49]张雄伟,陈亮等.现代语音处理技术及其应用[M].北京：机械工业出版社,2003
[50]李军林,杜松,崔慧娟,唐昆.0.8kb/s高质量声码器算法[J].清华大学学报(自然科学版),2003,43(1)：12-15
[51]Atal B.. Efficient Coding of LPC parameters by temporal decomposition[J]. Proc. of ICASSP, 1983:81-85
[52]K. K. Paliwal, W. B. Kleijn. Quantization of LPC Parameters in Speech Coding and Synthesis [J]. Elsevier Science B.V.,1995(12):433-466
[53]居来提.语音压缩编码[J].新疆教育学运学报,2002,18(2)：96-100
[54]张雄伟.2.4 kb/s MELP低速语音编码算法中的关键技术[J].电子工程师,1999(4)：21-23
[55]李文化.基于MELP的低速率语音编码算法研究[D].南京：南京邮电大学,2006
[56]程万灵.基于多带激励的低速率语音编码研究与实现[D].西安：西安电子科技大学,2009
[2]王洪,唐凯.低速率语音编码[M].北京：国防工业出版社,2005：122-148
[3]戚银城,张巍,苑津莎.超低速率MELP语音编码算法研究[J].声学技术,2007,26(6)：1196-1200
[4]雒国成,蔡芳.基于MELP算法的语音编解码器分析[J],黄河水利职业技术学院学报,2008,20(1)：49-51
[5]张毓忠,梅武钢.语音信号MELP编码的研究与实现[J].电声技术,2000(11)：6-9
[6]马震,蒋保臣,陶立慧.混合激励线性预测(MELP)方法研究与改进[J].现代电子技术,2005(13)：22-25
[7]陈双燕,张铁军,王东辉,侯朝焕.基于一种可配置可扩展处理器的MELP语音算法的改进与实现[J].微电子学与计算机,2006,23(6)：42-44
[8]王炳锡.语音编码[M].西安：西安电子科技人学出版社,2002：257-274
[9]徐贵森,刘鑫,潭学治.浅谈无线对讲机模拟转数字[J].移动通信,2007：22-25
[10]李慧芳.基于MELP的低码率语音编码技术研究与实现[D].北京：北京邮电大学,2008
[11]高浩.基于DSP关于改进的MBE语音算法的研究[D].江苏：江南大学,2006年
[12]王宇寰.基于声码器的900MHz数字对讲机研究[D].西安：西安电子科技大学,2007
[13]李娜,王中元,胡瑞敏,朱福喜.数字语音编码技术和标准介绍[J].中国新通信(技术版),2007(13)：3943
[14]程飞.基于MELP的高效声码器研究[D].北京：国防科学技术大学,2006
[15]史水平,李世作.线性预测编码技术及其在音频文件上的应用[J].现代电子技术,2004(4)：21-23
[16]洪治.我国对讲机市场现状及发展趋势[J].中国无线电,2007(1)：37,64
[17]刘雁飞.数字无中心多信道选址移动通信协议的研究[D].北京：北京化工大学,2008
[18]ETSITS 102 361-1. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR)Systems; Part1:DMR Air Interface (AI) protocol"[S].2007
[19]ETSI TS 102 361-2. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part2:DMR voice and generic services and facilities" [S].2007
[20]ETSI TS 102 361-3. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part 3:DMR Data protocol"[S].2007
[21]ETSI TS 102 361-4. "Electromagnetic compatibility and Radio spectrum Matters(ERM); Digital Mobile Radio(DMR) Systems; Part 4:DMR trunking protocol"[S].2007
[22]李进良.对制订我国数字专业移动通信标准的建议[J].移动通信,2008,17：9-13
[23]王保坤.DMR端机基带模块研究与实现[D].西安：西安电子科技大学,2010
[24]百度百科.多址方式[EB/OL]. http://baike.baidu.com/view/184175.htm
[25]李志芳.数字对讲机——对讲机技术发展的必然方向[J].现代通信,2009
[26]刘辉.DMR数字对讲机协议的研究与实现[D].北京：北京化工大学,2010
[27]ETSI TS 300 168. "Radio Equipment and Systems (RES); Digital Short Range Radio (DSRR)" [S].1993
[28]Alan V. McCree, Thomas P. Barnwell Ⅲ. A Mixed Excitation LPC Vocoder Model for Low Bit Rate Speech Coding[J]. IEEE Transaction On speech and audio processing,1995,3(4):242-250
[29]姚天任.数字语音处理[M].武汉：华中科技大学出版社,2007
[30]张旭.多带激励(MBE)低速率语音编码技术的研究与实现[D].天津：天津大学,2005
[31]易克初.语音信号处理[M].北京：国防工业出版社,2000：15-115
[32]刘媛媛.基于DSP的MELP语音编码算法研究及实现[D].哈尔滨：哈尔滨工程大学,2008
[33]杨行峻,迟惠生.语音信号数字处理[M].北京：电子工业出版社,1995：48-127
[34]Makhoul J. I. Linear Prediction:A Tutorial Review[J]. Proc. IEEE,1995,63(4):561-580.
[35]张毓忠,梅武钢.语音信号MELP编码的研究与实现[J].电声技术,2000(11)：6-9
[36]马震,蒋保臣,陶立慧.混合激励线性预测(MELP)方法研究与改进[J].现代电子技术,2005(13)：22-25
[37]戚银城,张巍,苑津莎.超低速率MELP语音编码算法研究[J].声学技术,2007(26)：1196-1200
[38]刘宽海,葛万成.MELP低比特率数字语音编码技术研究[J].信息技术,2003,27(8)：52-54
[39]张雄伟.2.4kb/s MELP低速率语音编码算法中的关键技术[J].电子工程师,1999(4)：21-23
[40 T. Tremain. The government standard linear Predictive coding algorithm:LPC-10[J]. Speech Technology Magazine,1982:40-49
[41]冷丰麟.基于声码化技术的数字语音加密系统的研究[D].大连：大连海事大学,2009
[42]D. W. Griffin. Multi-band excitation vocoder[J]. IEEE Transaction on ASSP,1988,36(8): 1223-1235
[43]D. W. Griffin, J. S. Lim. Signal estimation from modified short-time fourier transform[J]. IEEE Transaction On ASSP,1984,32(2):236-243
[44]P. C. Meuse. A 2400 bps multi-band excitation speech coder[J]. ICASSP,1990(88):9-12
[45]Griffith, D.W., Lim, J.S.. Multiband excitation vocoder[J]. IEEE Transactions on ASSP,1988, 36(8):1223-1235
[46]Alan V. McCree, E. Bryan George, Thomas P. Barnwell. A 2.4 kbit/s MELP coder candidate for the new U. S. Federal Standard[J]. Proceeding for ICASSP Atlanta,1996,1(1):200-203
[47]王都生.多带混合激励低速率语音编码的研究[D].西安：西安电子科技大学,1998
[48]胡剑凌,徐盛,陈健.2.4kb/sMELP算法设计[J].上海交通大学学报,2000,34(6)：789-792
[49]张雄伟,陈亮等.现代语音处理技术及其应用[M].北京：机械工业出版社,2003
[50]李军林,杜松,崔慧娟,唐昆.0.8kb/s高质量声码器算法[J].清华大学学报(自然科学版),2003,43(1)：12-15
[51]Atal B.. Efficient Coding of LPC parameters by temporal decomposition[J]. Proc. of ICASSP, 1983:81-85
[52]K. K. Paliwal, W. B. Kleijn. Quantization of LPC Parameters in Speech Coding and Synthesis [J]. Elsevier Science B.V.,1995(12):433-466
[53]居来提.语音压缩编码[J].新疆教育学运学报,2002,18(2)：96-100
[54]张雄伟.2.4 kb/s MELP低速语音编码算法中的关键技术[J].电子工程师,1999(4)：21-23
[55]李文化.基于MELP的低速率语音编码算法研究[D].南京：南京邮电大学,2006
[56]程万灵.基于多带激励的低速率语音编码研究与实现[D].西安：西安电子科技大学,2009