Triplet entropy loss: improving the generalisation of short speech language identification systems
Spoken language identification systems form an integral part in many speech recognition tools today. Over the years many techniques have been used to identify the language spoken, given just the audio input, but in recent years the trend has been to use end to end deep learning systems. Most of thes...
Main Author: | Van Der Merwe, Ruan Henry |
---|---|
Other Authors: | Er, Sebnem |
Format: | Dissertation |
Language: | English |
Published: |
Faculty of Science
2021
|
Subjects: | |
Online Access: | http://hdl.handle.net/11427/33953 |
Similar Items
-
Generalised bootstrap procedures
by: Lee, Stephen Man Sing
Published: (1993) -
Person Re-Identification With Triplet Focal Loss
by: Shizhou Zhang, et al.
Published: (2018-01-01) -
Generalisations of Fisher Matrices
by: Alan Heavens
Published: (2016-06-01) -
Person Re-Identification by Pose Invariant Deep Metric Learning With Improved Triplet Loss
by: Min Chen, et al.
Published: (2018-01-01) -
ON-BOARD GCPS MATCHING WITH IMPROVED TRIPLET LOSS FUNCTION
by: G. Xie, et al.
Published: (2020-08-01)