Location:
Search - speech MAP
Search list
Description: speech recognition speaker adaption map algorithm
Platform: |
Size: 1024 |
Author: lifei |
Hits:
Description: 语音识别中的说话人自适应研究.nh
1.MAP和MLLR算法比较
文章在讨论由说话人引起的声学差异基础上，研究两种基于模型
的自适应算法:最大似然线性回归(州压LR)和最大后验概率(MAp)。
实验结果表明，不论采用哪种自适应都能使识别率有一定的提升。两
种算法之间的差异性在于MAP具有良好的渐进性，但收敛性较差，
而MLLR在很大程度上改善了收敛特性，但其渐进特性却不如MAP。
文章讨论了在侧汰P自适应中，初始模型参数的先验知识对自适
应效果的影响，以及在MLLR中，回归类对自适应效果的影响。文
章还进一步研究了采用两种算法的累加自适应效果，从结果看MAP
和MLLR结合的方法比单独使用M[AP和MLLR的效果要好。文章
还对包括基于特征层的归一化算法和用于基于声学模型的MLLR算
法等效性进行讨论，并给出了统一的算法框架。-speech paper，help you study
Platform: |
Size: 5208064 |
Author: 海豚 |
Hits:
Description: 小车程序，很有用的，可以直接烧进去控制小车-A control system for the intelligent toy car is designed with SPCE061A MCU．The car is under control of speech command．The design and implementation of its hardware is discussed in detail. The Design Complete the power circuit, reset circuit, keyboard circuit, audio input circuit
Platform: |
Size: 3072 |
Author: 陈述 |
Hits:
Description: ：为了使应力变异在顽健语音识别系统中能够达到较好的识别效果，研究了基于隐马
尔可夫模型(HMM)的自适应技术，提出了将最大后验概率(MAP)和最大似然回归方法(MLLR)用
于应力变异语音的自适应中。实验结果表明，与基本系统相比，两种方法均有效地提高系统识别
率。以SD为初始模型的最大后验概率方法在150个训练样本时识别效果最好，可以达到90．4％ 。-: In order to stress variation in the robustness of speech recognition system can achieve better recognition results, based on Hidden Markov Model (HMM) of adaptive technology, put forward a maximum a posteriori probability (MAP) and Maximum Likelihood regression (MLLR) for the stress of the adaptive variation in voice. The experimental results show that compared with the basic system, both methods are effective to improve the system recognition rate. SD as the initial model to the maximum a posteriori probability method in 150 training samples to identify the best, can reach 90.4 .
Platform: |
Size: 234496 |
Author: 尹江波 |
Hits:
Description: 观测实时模拟信号（语音）的频谱
用音频设备采集一段语音，将语音存为.wav格式。对wav文件作分段傅里叶变换分析。语音是分音节的，应把它分段分析，而且实际运用中的数字信号处理的FFT的点数是有限的，一般只能达到千点。用傅里叶反变换IFFT，从频域恢复信号。画出频谱图和语音波形图。
-Observing real-time analog signal (voice) of the spectrum collected with the audio devices section of voice, voice saved as. Wav format. Wav file of the sub-Fourier transform analysis to make. Speech is divided into syllables, it should be sub-analysis, and practical application of digital signal processing of the FFT points is limited, generally only reach 1000 points. Fourier inverse transform IFFT, to restore the signal from the frequency domain. Draw the spectrum map and voice waveform.
Platform: |
Size: 63488 |
Author: benny |
Hits:
Description: 语音识别自适应训练原理的论文，包括MLLR，MAP-adaptation for speech recognition
Platform: |
Size: 987136 |
Author: 梁伟文 |
Hits:
Description: Based on the spectrum of the vuvuzela sound, this denoising technique simply computes an antenuation map in the time-frequency domain. Then, the audio signal is restored by computing the inverse STFT. See [1] or [2] for more detail about the algorithm.
To hear the result of this algorithm, go directly to: http://soundcloud.com/choc29/vuvuzela-correction-with-matlab
References:
[1] Steven F. Boll, "Suppression of Acoustic Noise in Speech Using Spectral Subtraction", IEEE Transactions on Signal Processing, 27(2),pp 113-120, 1979
Platform: |
Size: 1156096 |
Author: naarkhoo |
Hits:
Description: 说话人自适应是提高非特定人语音识别系统识别性能的有效手段，本文针对非母语说话人，结合常规的白适应技术MLLR和MAP，探索云南纳西族和傈僳族两种母语说话人的汉语普通话语音识别问题，实验结果显示有显著效果。-Speaker adaptation is a powerful means of improving the performance of speakerindependent speech recognition system．Aimed at Yunnan minority Naxi and Lisu speech，non—
native mandarin speech recognition is discussed aplflying general speaker adaptation MLLR and MAP．It is proved to be effective by the experiment results．
Platform: |
Size: 363520 |
Author: 自然快乐 |
Hits:
Description: speech recognision is used to develop automatic speech regonision system and also to normalise it and then map it to a particular user voice.
Platform: |
Size: 40960 |
Author: anusha123 |
Hits:
Description: 绘制出语音信号中一帧浊音的频谱，可以看出是具有谐波结构的频谱-Map out the spectrum of a voiced speech signal...
Platform: |
Size: 199680 |
Author: huzhengzheng |
Hits:
Description: 数字电子技术基础之逻辑代数基础讲稿，包含与或非，逻辑运算基本公式，卡诺图等-Fundamentals of Digital Electronics logic algebraic basis of speech, including with non-basic formula of the logic operations, Karnaugh map
Platform: |
Size: 815104 |
Author: 密蒙花开 |
Hits:
Description: 1.用matlab实现对一个元音和一个清辅音的LPC系数，LPCC系数和MFCC系数的提取。
2，用LPC对这两个语音信号的功率谱进行估计，其中，LPC系数要分别有5阶，10阶，15阶和20阶四种情况，并在一个图里画出信号本身的FFT功率谱和四条不同阶数的LPC谱估计图，以作比较。注意，LPC和LPCC只考察自相关法。
3，用对数面积比的方法来求出这两个发音的变截面声管模拟图形，10阶即可-Using matlab to achieve the LPC coefficients of a vowel and a voiceless consonant, the extraction of LPCC Features coefficients and MFCC coefficients. LPC both the speech signal power spectrum estimate, which, LPC coefficients, respectively, five bands, 10 bands, 15 bands and 20 the first four cases, and draw a graph of the FFT power of the signal itself spectrum and four different orders of LPC spectral estimation map, for comparison. LPC and LPCC only examine the autocorrelation method. 3, the number of area ratio method to calculate the variable cross-section of the two pronunciation sound tube simulation graphics, 10 bands can be
Platform: |
Size: 72704 |
Author: houjam |
Hits:
Description: 国家集训队2006论文集
陈启峰：《“约制、放宽”方法在解题中的应用》
陈首元：《维护森林连通性——动态树》
冯威：《数与图的完美结合——浅析差分约束系统》
高逸涵：《对于一道题目的深入分析》
胡伟栋：《演讲的若干建议》
黄劲松：《贪婪的动态规划》
黄晓愉：《深度优先搜索问题的优化技巧》
贾由：《由图论算法浅析算法优化》
李天翼：《从特殊情况考虑》
龙凡：《一类猜数问题的研究》
汤泽：《浅析队列在一类单调性问题中的应用》
唐文斌：《“调整”思想在信息学中的应用》
汪晔：《信息学中的参考系与坐标系》
王栋：《浅析平面Voronoi图的构造及应用》
王赟：《Trie图的构建、活用与改进》
余远铭：《最短路算法及其应用》
俞鑫：《棋盘中的棋盘——浅谈棋盘的分割思想》
周戈林：《浅谈类比思想》
周以苏：《论反汇编在时间常数优化中的应用》
朱晨光：《基本数据结构在信息学竞赛中的应用》
朱泽园：《半平面交的新算法及其实用价值》
-Proceedings of the National Team 2006
Chen Qifeng: "" restraint, relaxed "approach in solving problems in the application."
First Yuan Chen: "Maintaining forest connectivity- Dynamic tree"
Feng Wei: "the number of perfection and graphs- Analyzing differential restraint system."
Manhattan Han: "For an in-depth analysis of the subject."
Hu Weidong: "Several suggestions speech."
Huang Jinsong: "Greed dynamic programming."
Huang Yu: "depth-first search optimization techniques problem"
Jia From: "On the graph algorithm optimization algorithm"
Li Tianyi: "From special circumstances to consider."
Dragon Fan: "guess the number of a class of problems."
Yuzawa: "Analysis of the queue in a class of problems in monotone"
Tang Wenbin: "" adjustment "thinking in the application of information science."
Wang Ye: "Information science and coordinate reference system"
Wang Dong: "Construction and Application Analysis of planar Voronoi diagram"
Wang Yun: "Building Trie map, utilization a
Platform: |
Size: 9092096 |
Author: sxy |
Hits:
Description: 用matlab读取一个语音信号，对其进行变调，回音灯各种处理的设计报告，含代码与仿真图-Reads a speech signal with matlab, its tone, design reports echo lamp various treatments, including code and simulation map
Platform: |
Size: 276480 |
Author: 林森 |
Hits:
Description: 最大似然（ML）准则和最大后验概率（MAP）准则，通过反复训练模板能有较高的识别率，语音信号的采集与处理，数字信号处理课设。- Maximum Likelihood (ML) criteria and maximum a posteriori (MAP) criterion, Through repeated training hYAWUCalate have higher recognition rate, Acquisition and Processing of the speech signal, digital signal processing class-based.
Platform: |
Size: 7168 |
Author: 乔荣涛 |
Hits: