About this document

Efficient Keyword Spotting in Luganda by aman is a document available to read on EtoBox.

This document discusses a method for efficient keyword spotting in low-resource languages, specifically targeting humanitarian relief efforts in Africa. The approach utilizes a convolutional neural network (CNN) trained on dynamic time warping (DTW) scores derived from a small set of isolated keywords and a larger corpus of untranscribed speech, achieving significant performance improvements over traditional methods. Experiments demonstrate that the CNN-DTW keyword spotter can operate effectively in low-res

Author

aman

Language

EN