VGU RESEARCH REPOSITORY
Please use this identifier to cite or link to this item:
https://epub.vgu.edu.vn/handle/dlibvgu/572| Title: | English speech recognition | Authors: | Nguyen Hoang Vinh Khang | Keywords: | English speech recognition | Issue Date: | 2015 | Publisher: | Vietnamese-German University | Abstract: | Linear predictive coding (LPC) is a digital signal processing method for pattern recognition such as image and speech recognition. In order to be recognized, speech signal has to be processed and extract the frequencies corresponding to the pronunciation of the speech. Based on the theory of LPC, a simple isolated word speech recognition system is developed which is capable of recognizing five vowel letters of the alphabet: a, e, i, o, u. However, the characteristic features are different across subjects. Therefore, a training database is required for the methods of recognition. Specifically, the analyzed recorded speech of a subject can be recognize if the subject's training data is included in the training database. During the development of the system, the recognition results are often wrong and the problem is identified as the difference between the pitches of male voice and female voice. Therefore, a further research for an algorithm called simple inverse filter tracking (SIFT) pitch detector is required for identifying the gender of speakers. Overall, the speech recognition system using LPC can automatically identify the gender of speakers and recognize the content of their speech. |
URI(1): | http://epub.vgu.edu.vn/handle/dlibvgu/572 | Rights: | Attribution-NonCommercial 4.0 International |
| Appears in Collections: | Electrical Engineering and Information Technology (EEIT) |
Files in This Item:
| File | Description | Size | Format | Existing users please Login |
|---|---|---|---|---|
| English speech recognition.pdf | 17.45 MB | Adobe PDF |
Page view(s)
92
checked on Nov 16, 2025
Download(s)
20
checked on Nov 16, 2025
Google ScholarTM
Check
This item is licensed under a Creative Commons License