【作者】 赵鹏；

【导师】 苏娟；

【作者基本信息】 湖南大学 ， 电路与系统， 2006， 硕士

【摘要】 为了克服传统汉语数码语音识别系统抗噪性差、识别率低的特点,本文阐述了一种基于TMS320VC5402定点数字信号处理器(DSP)的连接汉语数码语音识别系统的设计和实践,力争使系统具有实时性、较强抗噪性、较高识别率和非特定人连接数码语音识别的特点。针对传统的“改进谱相减法语音增强”参数设定单一、环境适应能力差的缺点,提出了一种利用模糊理论和“改进的谱相减法”结合的“模糊谱相减法语音增强”;针对语音信号端点检测困难的特点,通过MATLAB仿真试验,给出了能够准确确定数码语音端点的初始和改进参数表;提出了利用基于线性预测编码倒谱参数和差分线性预测编码倒谱参数相结合的离散隐含马尔可夫模型进行第一级识别、利用共振峰参数进行第二级识别的两级汉语数码语音识别系统,在保证系统实时性的同时,实现连接汉语数码语音识别系统识别率的提高;在硬件实现上,详细阐述了基于TMS320VC5402的连接汉语数码语音识别系统各部分硬件设计;在软件开发上,给出了连接汉语数码语音识别的软件设计各部分的流程图,并对各部分进行了MATLAB仿真,并给出了仿真结果。最后,分别建立了数码语音识别仿真系统和连接数码语音训练系统。利用连接数码语音训练系统得到了男女各一套向量量化码本和男女各一套11个数码的非特定人连接数码语音离散隐含马尔可夫参数;基于这些参数,连接数码语音识别仿真系统成功实现了对输入数码语音的识别,并且系统具有较好的抗噪性。更多还原

【Abstract】 In order to overcome the disadvantages of traditional Mandarin Digit Speech Recognition System, including bad robust and low recognition rate, this thesis elaborates the theory and practice of design of Connected Mandarin Digit Speech Recognition (CMDSR) system based on TMS320VC5402 fixed point Digital Signal Processor (DSP). This thesis tries to update the CMDSR system to achieve the characters below: real-time, better robust, higher recognition rate, non-special-man. Considering the disadvantages of traditional Improved Spectrum Subtraction Speech Enhancement, this thesis proposes the theory of Fuzzy Spectrum Subtraction based on the Fuzzy Theory and Improved Spectrum Subtraction Speech Enhancement; as for the difficulties of detecting the endpoint of speech signal, the thesis gives the table of initial and the improved parameters, with which we can confirm the endpoints of Mandarin Digit Speech; the thesis puts forward two-level digit real-time speech recognition system, the first level is based on Discrete Hidden Markov Model which is Linear Predictive Coding Cepstrum (LPCC) and Difference Linear Predictive Coding Cepstrum (DLPCC) , the second level is based on formant parameters; as for the realization of hardware, the thesis depicts the realization of every part of CMDSR based on the TMS320VC5402 in detail; as for the development of software, the thesis gives the software design flow chart of CMDSR, simulates the basic theory with MATLAB language and gives the simulation results.At last, the thesis establishes Mandarin Digit Speech Recognition Simulation System (MDSRSS) and Connected Mandarin Digit Speech Training System (CMDSTS) separately; with the CMDSTS, the thesis gets two sets of Vector Quantization (VQ) parameter table, including man’s and woman’s, besides, it gets two sets of non-special-man CMDSR Discrete Hidden Markov Model (DHMM) parameters, including man’s and woman’s as well. With the tables and parameters, the MDSRSS can recognize the input digit speech successfully and it also has better robust character.更多还原