No self-introduction
Loading
Profile
Follow
codes (6)
Speaker recognition based on LPCC and MFCC
3.7
MFCC parameters and speaker recognition LPCC parameters are the two most commonly used features of the parameters studied algorithm principle and LPCC MFCC parameter extraction and poor Points cepstrum parameter extraction method, using MFCC, LPCC and the first order, second order difference as the characteristic parameter by k-means algorithm and the BP neural network to say Word recognition. The experimental results show that this method can effectively improve the recognition rate, but also to verify the robustness of MFCC parameters than LPCC parameters
duandian 0 0
1
HMM model speech recognition source code
4.2
The speech recognition system based on CHMM model can well realize the function of Chinese speech recognition.
duandian 12 0
1
Student achievement management system
no vote
Query by student number, gender and class. The query method can be supplemented by itself. &You can count the average score and total score of each student, and sort them according to the average score and total score (there are many sorting methods, so you should check the sorting methods on the Internet and compare the quality of each sorting method), and print the sorting results on the screen. &It can modify, delete and add students' information. Each function module of the system needs to be realized in the form of function. In the main function through the menu to call each function. &Use linked list or array to store class students.
duandian 0 0
1
HMM training algorithm
no vote
Hidden Markov models (HM m) traditional training methods of--Baum-W Elch algorithm can only be a local optimum model, thus affecting the
Final recognition rate. CHM m, subparagraph k average method to obtain the original value can solve this problem, but the DHM m improvements are unlikely.
An important feature of evolutionary computation based on global search can be suboptimal solutions, as well as the global optimal solution. This article introduced into evolutionary computation
DHM m in training, and evolution of an improved training method and experimental results indicate that this training approach in its global search and fast convergence
Characteristics of the resulting model is superior to the traditional method and directly derived from evolutionary com
duandian 2 0
1
Speech recognition models Hmm
no vote
On non-specific people of voice recognition code, with c language prepared, can transplant to various embedded platform, used HMM model, contains Qian to, Hou to, Victor than algorithm, detailed code, run through, can with, for beginners, helps voice recognition of development........................................
duandian 1 0
1
Speech recognition of master's thesis
no vote
First, the speech signal pre-processing and character extraction issues were discussed, and extracted three valid language Characteristic parameters of sound recognition, LPC-cepstrum coefficients-LPC coefficients and the Mel-frequency Cepstral coefficients (MFCC), followed by smoked discusses two commonly used in speech recognition and recognition-DTW based on template matching based on statistics of membrane-type application of HMM in speech recognition, discussed separately individual models of training and recognition
Law of HMM in the practical application of some specific issues have also been discussed; Finally, conventional HMM Shows some improvement methods, including explicitly state presence and the random segmentation model (SM), focusing on SM
Training in speech re
duandian 0 0
1
No more~
Follows0
Fans0