 |
Conference papers
AN OPEN-SOURCE SPEAKER GENDER DETECTION FRAMEWORK FOR MONITORING GENDER EQUALITY
Abstract : This paper presents an approach based on acoustic analysis to describe gender equality in French audiovisual streams, through the estimation of male and female speaking time. Gender detection systems based on Gaussian Mixture Models , i-vectors and Convolutional Neural Networks (CNN) were trained using an internal database of 2,284 French speakers and evaluated using REPERE challenge corpus. The CNN system obtained the best performance with a frame-level gender detection F-measure of 96.52 and a hourly gender speaking time percentage error bellow 0.6%. It was considered reliable enough to realize large-scale gender equality descriptions. The proposed gender detection system has been packaged as an open-source framework.
Cited literature [28 references]
https://hal.archives-ouvertes.fr/hal-01927560
Contributor : Anthony Larcher <>
Submitted on : Monday, November 19, 2018 - 10:57:53 PM
Last modification on : Sunday, May 2, 2021 - 3:30:28 AM
Long-term archiving on: : Wednesday, February 20, 2019 - 4:23:33 PM
Contributor : Anthony Larcher <>
Submitted on : Monday, November 19, 2018 - 10:57:53 PM
Last modification on : Sunday, May 2, 2021 - 3:30:28 AM
Long-term archiving on: : Wednesday, February 20, 2019 - 4:23:33 PM
File
ddoukhan_icassp_2018.pdf
Files produced by the author(s)