Abstract
The sound recognition is considered as an effective tool to improve the performance of man-machine interaction. However, because of the non-ideal effect during the propagation and the extensive variation range of sound signal, it is quite difficult to achieve high accuracy target sound recognition for conventional sound recognition methods. In order to improve the performance of sound recognition, a robust sound recognition method based on human auditory bionic processing is proposed in this paper. In the proposed method, by analyzing the major function of human auditory system, the mathematical model of human auditory system is firstly developed to simulate the sound propagation in human auditory physiological processing. Then by extracting the muti-dimensional eigenvectors of the auditory spectrum, the feature of target sound signal is obtained. Afterwards, the recognition and classification of sound signal is achieved by the back-propagation (BP) neural network method. Based on the simulation of actual sound signal, it is verified that the proposed method can effectively simulate the sound propagation and achieve desirable sound recognition performance under noise condition.
| Original language | English |
|---|---|
| Publication status | Published - 2015 |
| Event | IET International Radar Conference 2015 - Hangzhou, China Duration: 14 Oct 2015 → 16 Oct 2015 |
Conference
| Conference | IET International Radar Conference 2015 |
|---|---|
| Country/Territory | China |
| City | Hangzhou |
| Period | 14/10/15 → 16/10/15 |
Keywords
- Back propagation neural network
- Feature extraction
- Human auditory system
- Muti-dimensional eigenvectors
- Sound recognition