Enhanced Speaker Embeddings with CNN-based Multi-scaling Features.

Maros Jakubec Roman Jarina Eva Lieskovska

Published in: RADIOELEKTRONIKA (2022)

Keyphrases

feature extraction
image features
cellular neural networks
feature vectors
low level
vector space
benchmark datasets
hidden markov models
extracting features
visual features
multi modal
machine learning
high dimensional
support vector
similarity measure
image processing
computer vision
information retrieval