ISCA Archive MAVEBA 2005
ISCA Archive MAVEBA 2005

Evaluation of speaker normalization for suicidality assessment

Khazaimatol S. Subari, D. Mitchell Wilkes, Stephen E. Silverman, Marilyn K. Silverman, Richard G. Shiavi

When reviewing his clinical experience in treating suicidal patients, one of the authors observed that successful predictions of suicidality were often based on the patient’s voice independent of content. Using the Gaussian mixture model to represent the mel-cepstral features of voiced speech, speech of suicidal persons can be distinguish from that of depressed and control persons. The question then becomes can warping of the frequency axis improve the classification. The results show that warping of the frequency axis using the third format or Gaussian mixture model technique produces the best classification results.

Index Terms. Speech, suicide, mel-cepstrum, frequency warping, classification


Cite as: Subari, K.S., Wilkes, D.M., Silverman, S.E., Silverman, M.K., Shiavi, R.G. (2005) Evaluation of speaker normalization for suicidality assessment. Proc. Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2005), 181-184

@inproceedings{subari05_maveba,
  author={Khazaimatol S. Subari and D. Mitchell Wilkes and Stephen E. Silverman and Marilyn K. Silverman and Richard G. Shiavi},
  title={{Evaluation of speaker normalization for suicidality assessment}},
  year=2005,
  booktitle={Proc. Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2005)},
  pages={181--184}
}