Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SP-L5.1 |
| Session: | Pitch and Tone Based Speech Analysis |
| Time: | Thursday, May 20, 09:30 - 09:50 |
| Presentation: |
Lecture |
| Topic: |
Speech Processing: Speech Analysis |
| Title: |
PITCH PREDICTION FROM MFCC VECTORS FOR SPEECH RECONSTRUCTION |
| Authors: |
Xu Shao; University of East Anglia | | |
| | Ben Milner; University of East Anglia | | |
| Abstract: |
This work proposes a technique for reconstructing an acoustic speech signal solely from a stream of mel-frequency cepstral coefficients (MFCCs). Previous speech reconstruction methods have required an additional pitch element, but this work proposes two maximum a posteriori (MAP) methods for predicting pitch from the MFCC vectors themselves. The first method is based on a Gaussian mixture model (GMM) while the second scheme utilises the temporal correlation available from a hidden Markov model (HMM) framework. A formal measurement of both frame classification accuracy and RMS pitch error shows that an HMM-based scheme with 5 clusters per state is able to correctly classify over 94% of frames and has an RMS pitch error of 3.1Hz in comparison to a reference pitch. Informal listening tests and analysis of spectrograms reveals that speech reconstructed solely from the MFCC vectors is almost indistinguishable from that using the reference pitch. |
| |
| Back | |