Hybrid Feature Selection Method Based on Harmony Search and Naked Mole-Rat Algorithms for Spoken Language Identification From Audio Signals.
Samarpan GuhaAankit DasPawan Kumar SinghAli AhmadianNorazak SenuRam SarkarPublished in: IEEE Access (2020)
Keyphrases
- feature selection
- unsupervised learning
- optimization method
- optimization algorithm
- harmony search
- computational complexity
- genetic algorithm
- language identification
- feature set
- optimization problems
- audio signals
- neural network
- bp neural network
- speech recognition
- model selection
- probabilistic model
- low level
- evolutionary algorithm
- similarity measure
- information retrieval
- machine learning