Segmented Time-Frequency Masking Algorithm for Speech Separation Based on Deep Neural Networks.
Xinyu GuoShifeng OuMeng GaoYing GaoPublished in: CISP-BMEI (2020)
Keyphrases
- experimental evaluation
- neural network
- computational cost
- preprocessing
- times faster
- np hard
- k means
- computational complexity
- pattern recognition
- learning algorithm
- high accuracy
- multiresolution
- cost function
- matching algorithm
- segmentation algorithm
- optimization algorithm
- detection algorithm
- training algorithm
- convergence rate
- frequency domain
- simulated annealing
- fuzzy logic
- wavelet transform
- optimal solution
- genetic algorithm