Description: 基于GMM的概率神经网络PNN具有良好的泛化能力，快速的学习能力，易于在线更新,并具有统计学的贝叶斯估计理论基础，已成为一种解决像说话人识别、文字识别、医疗图像识别、卫星云图识别等许多实际困难分类问题的很有效的工具。而且PNN不但具有GMM的大部分优点，还具有许多GMM没有的优点，如强鲁棒性，需要更少的训练语料，可以和其他网络其他理论无缝整合等。-GMM based probabilistic neural network PNN good generalization ability, the ability to learn fast, easy online updates, and with the Bayesian statistical theory based on estimates, and has become a solution as speaker recognition, text recognition, medical image recognition, satellite images and other real recognition when difficulties classification of very effective tool. But GMM PNN is not only the most advantages, but also has many advantages GMM not as strong robustness, require less training corpus, and other networks to other theories, such as seamless integration. Platform: |
Size: 7158 |
Author:姜正茂 |
Hits:
Description: 在MATLAB环境下实现基于矢量量化的说话人识别系统。在实时录音的情况下，利用该说话人识别系统，对不同的人的1s～7s的语音进行辨识。实现与文本无关的自动说话人确认的实时识别。
使用说明：
1 训练
打开Matlab 使Current Directory为VQ所在的文件夹(比如：E:\\vq)
在Command windows中输入
-in MATLAB environment based VQ Speaker Recognition System. Real-time recording in the circumstances, the speaker recognition system, the different people of one's voice-7s for identification. Implementation and text-independent speaker recognition automatic real-time identification. Use : Matlab opened a training Current Directory to make VQ where the folder (for example, : E : \\ vq) in the Command windows input Platform: |
Size: 707106 |
Author:annya |
Hits:
Description: 关于说话人识别方面的五个子程序，包括倒谱基音周期混合特征系数的话者识别，能频积端点检测、语音基音周期检测等C++源代码，本人整理编译过，比较紧凑高效；-Speaker Recognition of the five subroutines, including cepstrum Pitch mixed coefficient of the recognition, the frequency can plot endpoint detection, Pitch detection C source code, I compiled finishing more compact and efficient; Platform: |
Size: 3332258 |
Author:kingvi |
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: 23925 |
Author:张开 |
Hits:
Description: 对音频文件提取其MFCC特征参数
可用于说话人识别-right audio files from their MFCC parameters can be used to Speaker Recognition Platform: |
Size: 515342 |
Author:康丹丹 |
Hits:
Description: An Evaluation of Visual Speech Features for the Tasks of Speech and Speaker Recognition.-An Evaluation of Visual Speech Features az r Tasks of the Speech and Speaker Recognition. Platform: |
Size: 102080 |
Author:rw |
Hits:
Description: 用三层小波神经网络实现的与文本无关说话人识别。（识别部分）。输入的是语音特征，输出的是识别结果。训练用的语音特征要事先提取出。-with three wavelet neural network has nothing to do with the text of Speaker Recognition. (Recognition). The admission of voice features, the output is the result of recognition. Training voice features prior to extract. Platform: |
Size: 6193 |
Author:白莹 |
Hits:
Description: 用c语言编写的语音特征参数的程序，该语音特征参数为mfcc参数，主要用于说话人识别-used language voice characteristic parameters of the procedure, the characteristic parameters for voice mfcc parameters, mainly for Speaker Recognition Platform: |
Size: 3081 |
Author:张龙 |
Hits:
Description: 用三层小波神经网络实现的与文本无关说话人识别。（识别部分）。输入的是语音特征，输出的是识别结果。训练用的语音特征要事先提取出。-with three wavelet neural network has nothing to do with the text of Speaker Recognition. (Recognition). The admission of voice features, the output is the result of recognition. Training voice features prior to extract. Platform: |
Size: 6144 |
Author:白莹 |
Hits:
Description: 基于GMM的概率神经网络PNN具有良好的泛化能力，快速的学习能力，易于在线更新,并具有统计学的贝叶斯估计理论基础，已成为一种解决像说话人识别、文字识别、医疗图像识别、卫星云图识别等许多实际困难分类问题的很有效的工具。而且PNN不但具有GMM的大部分优点，还具有许多GMM没有的优点，如强鲁棒性，需要更少的训练语料，可以和其他网络其他理论无缝整合等。-GMM based probabilistic neural network PNN good generalization ability, the ability to learn fast, easy online updates, and with the Bayesian statistical theory based on estimates, and has become a solution as speaker recognition, text recognition, medical image recognition, satellite images and other real recognition when difficulties classification of very effective tool. But GMM PNN is not only the most advantages, but also has many advantages GMM not as strong robustness, require less training corpus, and other networks to other theories, such as seamless integration. Platform: |
Size: 7168 |
Author:姜正茂 |
Hits:
Description: This class library contains more than 20 classes including feature extraction algorithms (MFCC, LPCC) and modeling techniques (HMM, GMM, DTW, VQ ) for automatic speech recognition and speaker verification Platform: |
Size: 1405952 |
Author:站长 |
Hits:
Description: 动态时间规整（DTW）算法基于动态规划的思想，解决了说话人不同时期发音长短、语速不一样的匹配问题。-dynamic time warping (DTW) algorithm is based on dynamic programming ideas, Speaker solve the different pronunciation of the length of time, Ma'am not match the same problem. Platform: |
Size: 417792 |
Author:陈宗耿 |
Hits:
Description: 基于MATLAB实现的DWT+GMM说话人识别，附带语料库和相关论文word版，毕设作品，详细且可运行！(MATLAB based DWT+GMM speaker recognition, incidental corpus and related papers word version, set up work, detailed and run!) Platform: |
Size: 11159552 |
Author:vanirgod |
Hits:
Search - speaker recognition