.png){kind=icon}
.png){kind=icon}
Return
Enhancing MFCC Feature Extraction Through Reservoir Computing
.png){kind=icon}
PRE
AI
DOI：10.1007/978-3-032-15641-9_20.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The extraction of features from speech is the most critical process in speech signal processing. Mel Frequency Cepstral Coefficients (MFCC) are the most widely used features in the majority of the speaker and speech recognition applications, as the filtering in this feature is similar to the filtering taking place in the human ear. But the current MFCC extraction is complex and requires time-frequency translations. Through our investigation, we were able to model a reservoir as a feature extractor capable of extracting the Mel Frequency Cepstral Coefficient (MFCC) without time-frequency domain translations. We have developed a real-time audio signal processing system by simplifying audio signal processing through the utilization of reservoir computers, which are significantly easier to train. We have established an experimental framework for end-to-end audio processing utilizing the reservoir and have investigated its capability to perform end-to-end audio signal processing.
Keywords:
Reservoir computing
Audio signal processing
MFCC
Journal
U
IF:
0
Papers:
27
Citations:
0