机器人语音识别技术研究

机器人语音识别技术研究(任务书,开题报告,论文14000字)
摘要
自机器人研究开始,机器人控制最简单,最方便的控制手段是语音控制。目前,从研究现状来看,机器人、机器人通信识别技术正逐渐成为研究的热点。语音识别技术使机器人能够将识别为语音激活信号的信息用于理解人类自然语言编号适用于各种机器人技术领域。将语音识别技术应用于机器人可以为用户带来便利。因此,实际机器人语音识别系统的研究和开发应该被广泛应用于机器人。这意义重大。论文的主要内容如下:
在语音识别的基础上,分析了机器人会话中语音识别的关键技术,并对语音信号进行预处理,包括采样、消噪;结束检测,预加重,窗口帧等。 Mel谱倒谱系数(MFCC)性能分析;研究了语音识别技术,主流模式训练和模式匹配技术动态时间规整(DTW)的核心。
最初的机器人语音识别和控制功能是通过matlab编程实现的。
关键词：机器人，语音识别，动态时间规整算法（DTW）

Abstract
Since the development of robots, voice control has been the most natural and convenient control for robot control System. From the perspective of the research status of voice recognition technology at home and abroad, communicate with robots and put voice Recognition technology applied to robots is becoming a hot topic of research.
   Speech recognition technology enables robots to understand the natural language of humans, using information identified as a voice-activated signal is applied to various technical fields of robots. Applying voice recognition technology to robots brings users Great convenience. Therefore, the research and development of a practical robot speech recognition system is widely used for robots. It is of great significance. The main content of the paper is as follows:
    First, based on the basic principles of speech recognition, the key to speech recognition for robot dialogue is studied.Technology, preprocessing of speech signals, including sampling, noise removal, endpoint detection, pre-emphasis, windowing separate frame and so on.; Performance of Linear Prediction Cepstrum Coefficient (LPCC) and Mel Frequency Cepstrum Coefficient (MFCC)Contrastive analysis; researched mainstream model training and pattern matching technology, which is the core of speech recognition technology Parts, including Hidden Markov Models (HMM), Dynamic Time Warping (DTW), Vector Quantization (VQ),Artificial neural network (ANN), a hybrid model of HMM and ANN, etc.
   The initial robot speech recognition and control functions are implemented by matlab programming.
Keywords: Robot，Speech Recognition，Dynamic Time Warping (DTW)

目录
摘要    I
Abstract    II
1. 绪论    1
1.1课题研究背景    1
1.2机器人概念    1
1.3语音识别    2
1.3.1语音概述    2
1.3.2语音技术    2
1.3.3语音技术的瓶颈    2
1.4国内外语音识别技术历史及其现状    2
1.4.1国外历史现状    2
1.4.2国内历史现状    4
1.5机器人语音识别技术的存在意义    5
2. 语音识别基本知识理论    5
2.1基本听觉语音学    5
2.2基本发音语音学    6
2.3基本声觉语音学    6
2.4基本汉语语音学    6
2.5基本激励模型与声道模型    7
2.5.1激励模型    7
2.5.2声道模型    7
2.5.3辐射模型    8
3. 语音识别关键技术    9
3.1整体原理及其框图    9
3.2语音信号的收集和预处理    9
3.2.1预滤波和采样    9
3.2.2去噪    9
3.2.3端点监测    10
3.2.4预加重    11
3.2.5加窗    11
3.3语音信号的转化以及特征提取    12
3.3.1 时域特征参数    12
3.3.2频域特征参数    12
3.4语音信号的模式匹配算法    14
3.5机器人语音识别的动作实现    15
4. 结论与展望    16
5. 参考文献    17
6. 致谢    18