Location:
Search - speech paper
Search list
Description:
Platform: |
Size: 17596416 |
Author: 方科颖 |
Hits:
Description: 说话人识别是语音识别的一种特殊方式，其目的不是识别语音内容，而是识别说话人是谁，即从语音信号中提取个人特征。采用矢量量化（VQ）可避免困难的语音分段问题和时间归整问题，且作为一种数据压缩手段可大大减少系统所需的数据存储量。本文提出了识别特征选取采用复倒谱特征参数和对应用VQ的说话人识别系统改进的一种方法。当用于训练的数据量较小时，复倒谱特征可以得到比较稳定的识别性能。VQ的改进方法避免了说话人识别系统的训练时间与使用时间相差过长从而导致系统的性能明显下降以及若利用自相关函数带来的大量运算。-Speaker Recognition Speech Recognition is a special way, its purpose is not voice recognition, Who identification but said that the voice signal from extracting personal characteristics. Vector quantization (VQ) can avoid the difficulties subparagraph voice to the issues and the whole time, and as a means of data compression system can significantly reduce the required data storage capacity. This paper presents a selection of identifiers employ Cepstrum parameters and the application of VQ speaker recognition system to improve a side France. When training for the amount of data is smaller, rehabilitation Cepstrum be relatively stable recognition performance. VQ improved ways to avoid the speech recognition system of training and the use of the difference in time, resulting in excessive sys
Platform: |
Size: 23552 |
Author: 张开 |
Hits:
Description: This paper deals with the problem of speech enhancement when a
corrupted speech signal with an additive colored noise is the only
information available for processing. Kalman filtering is known as
an effective speech enhancement technique, in which speech signal
is usually modeled as autoregressive (AR) process and represented
in the state-space domain.-This paper deals with the problem of speech enhancement when a corrupted speech signal wit h an additive colored noise is the only informat ion available for processing. Kalman filterin g is known as an effective speech enhancement te chnique. in which speech signal is usually modeled as aut oregressive (AR) process and represented in th e state-space domain.
Platform: |
Size: 102400 |
Author: rifer |
Hits:
Description: This paper deals with the problem of speech enhancement when
only a corrupted speech signal is available for processing. Kalman
filtering is known as an effective speech enhancement technique,
in which speech signal is usually modeled as autoregressive (AR)
model and represented in the state-space domain.-This paper deals with the problem of speech enhancement when only a corrupted speech signa l is available for processing. Kalman filterin g is known as an effective speech enhancement te chnique. in which speech signal is usually modeled as aut oregressive (AR) model and represented in the s tate-space domain.
Platform: |
Size: 245760 |
Author: rifer |
Hits:
Description: 语音识别的例子。请先到微软官方网站下载语音识别包和语音识别SDK，本程序根目录下有一个语法命令文件。-speech recognition examples. Microsoft advised to download the official website of packet voice recognition and voice recognition SDK, The procedures under the root directory is a grammatical order paper.
Platform: |
Size: 245760 |
Author: kun |
Hits:
Description: 用于语音情感识别软件中所需的声音文件,文件格式为MP3,包括悲伤,愤怒,高兴,惊恐,平静等声音资料.-for Emotional Speech Recognition Software for the voice paper for the MP3 format, including sadness, anger, happiness, Panic, calm voice, and other information.
Platform: |
Size: 118784 |
Author: beijingolympic.hi |
Hits:
Description: DTW算法在实现小词汇表孤立词识别系统时既简单又有效 ,在特定的场合下获得了广泛的应用。但DTW算法实
际应用时有许多缺点 ,本文对语音识别数学模型DTW作了深入的研究 ,提出了改进算法-DTW algorithm in small vocabulary isolated word recognition system is simple and effective, in particular occasions to obtain a wide range of applications. However, the practical application of DTW algorithm when there are many shortcomings, this paper, the mathematical model for speech recognition DTW-depth studies were made to improve the algorithm
Platform: |
Size: 216064 |
Author: 在式朝 |
Hits:
Description: Matlab是常用的演草纸式编程工具，好多非计算机专业的都可以掌握，若能利用它控制实现与单片机、DSP等下位计的通信，则可轻易实现控制与数据的采集，然后利用matlab强大的运算和处理功能进行后继的处理，这样对于研究生作毕业设计非常有利。利用Matlab实现控制串口通信，控制下位机，VC.m是起始界面，其他的为子程序，希望对大家有帮助。-Matlab is commonly used toilet paper speech programming tools, a lot of non-computer professionals can grasp, if the use of its control with single-chip, DSP wait for the next bit of communication, can easily realize the control and data collection, and then use matlab powerful computing and processing capabilities to carry out follow-up treatment, such as graduation project for graduate students is very beneficial. Realize the use of Matlab to control serial communications, under the control of digital machines, VC.m is the initial interface, the other for the subroutine, in the hope that everyone has to help.
Platform: |
Size: 11264 |
Author: 唐志 |
Hits:
Description: 基于DTW的孤立词语音识别研究和算法改进，很不错的论文-DTW-based isolated word speech recognition research and algorithm improvements, very good paper
Platform: |
Size: 154624 |
Author: 陈吉成 |
Hits:
Description: 本文针对语音识别处理系统中面临的主要问题和关键点，利用语音处
理系统对语音处理的一些关键点进行讨论研究。并且利用一些基本算法对语音
处理的部分参数进行实验。
-In this paper, speech recognition systems deal with major issues and key points, use of voice processing systems to handle voice some of the key points in my research. And use some of the basic algorithm to deal with part of speech parameters of the experiment.
Platform: |
Size: 2975744 |
Author: xcs |
Hits:
Description: 盲信号分离(BSS)是指在对彼此独立的源信号混合过程及各源信号本身均未知的情况下，从混合信号中分离出这些源信号的方法。BSS可以用来从多个麦克风混合语音信号中提炼出单个语音信号。本文简要阐述LMS、RLS算法，并通过仿真实验来分析比较这两类方法的性能，并利用此方法对一实际的语音信号进行分离。-Blind Signal Separation (BSS) is defined as independent of each other mixed-signal process and the source of the signal itself are unknown circumstances, from the mixed-signal to isolate the source of these signals. BSS can be used from multiple microphones mixed voice signal to extract a single speech signal. This paper briefly described LMS, RLS algorithm, and through simulation experiments to analyze the comparative performance of these two types of methods and take advantage of this method on a real speech signal separation.
Platform: |
Size: 1164288 |
Author: 贡晓飞 |
Hits:
Description: 本文主
要讨沦了语音检测的一些常用方法，并采用自相关方法进行了
模拟实验-This paper discussed the theory of some commonly used voice detection methods and the use of auto-correlation method simulation
Platform: |
Size: 334848 |
Author: 黄观杰 |
Hits:
Description: This paper is used in speech enhancement based on wavlet.
Platform: |
Size: 192512 |
Author: |
Hits:
Description: This paper is used in speech enhancement based on wavlet.
Platform: |
Size: 291840 |
Author: |
Hits:
Description: 本文介绍了用HMM嵌入训练方法建立连续语音的声学模型-In this paper, HMM embedded training methods used to establish continuous speech acoustic model
Platform: |
Size: 304128 |
Author: 杨絮 |
Hits:
Description: 《Neural Networks for Text-to-Speech Phoneme Recognition》This paper presents two different artificial neural network approaches for phoneme recognition for text-to-speech applications: Staged Backpropagation Neural Networks and Self-Organizing Maps.- Neural Networks for Text-to-Speech Phoneme Recognition This paper presents two different artificial neural network approaches for phoneme recognition for text-to-speech applications: Staged Backpropagation Neural Networks and Self-Organizing Maps.
Platform: |
Size: 722944 |
Author: 付诗 |
Hits:
Description: Based on short-term energy detection and short-term cross zero rates detection in speech
reorganization,the paper presents two-threshold endpoint detection.In addition,an accurate speech
segmentation algorithm is achieved with the wavelet transform and statistics signal frequency domain
characteristics.using cepstrum transform and cepstrum distance with the background noise,the paper achieved
the speech segmentation algorithm.Analysis and Implementation of Speech Endpoint Detection
Platform: |
Size: 413696 |
Author: 读宴宾 |
Hits:
Description: 清华大学关于语音端点检测的论文 清华大学关于语音端点检测的论文-Tsinghua University, on Speech Endpoint Detection Tsinghua University dissertation on the voice activity detection paper on Speech Endpoint Detection Tsinghua papers
Platform: |
Size: 751616 |
Author: appolo |
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: effecient speech recognition using wavelet and g-effecient speech recognition using wavelet and gmm
Platform: |
Size: 229376 |
Author: deepa |
Hits: