Digital Signal Processing Reference
In-Depth Information
contribution of different parts of the utterances toward emotion recognition is stud-
ied by developing emotion recognition models using the prosodic features obtained
from initial, middle, and final regions of the utterances. The combination of local
and global prosodic features was found to marginally improve the performance com-
pared to the performance of the systems developed using only local features. From
the word and syllable level prosodic analysis, the unique observation in the view of
discriminating the emotions is that final words and syllables contain more emotion
discriminative information than the other groups of words and syllables.
References
1. K. Murty, B. Yegnanarayana, Epoch extraction from speech signals. IEEE Trans. Audio Speech
Lang. Process.
16
, 1602-1613 (2008)
2. S.R.M. Prasanna, B.V.S. Reddy, P. Krishnamoorthy, Vowel onset point detection using source,
spectral peaks, and modulation spectrum energies. IEEE Trans Audio Speech Lang Process
17
,
556-565 (May 2009)
3. S.R.M. Prasanna, J.M. Zachariah, Detection of vowel onset point in speech, in
Proceedings
of IEEE International Conference Acoustics Speech, Signal Processing
Orlando, Florida, USA
May 2002
4. S.R.M. Prasanna,
Event-Based Analysis of Speech
. PhD thesis, Department of Computer Science
and Engineering, Indian Institute of Technology Madras, Chennai, India, Mar 2004
6. K.S. Rao, S.G. Koolagudi, R.R. Vempada, Emotion recognition from speech using global and
local prosodic features. Int. J. Speech Technol.
15
, 265-289 (2012). doi:
10.1007/s10772-012-
Search WWH ::
Custom Search