Direction of Arrival Estimation of Noisy Speech using Convolutional Recurrent Neural Networks with Higher-Order Ambisonics Signals.
Nils PoschadelRobert HupkeStephan PreihsJürgen PeissigPublished in: EUSIPCO (2021)
Keyphrases
- recurrent neural networks
- direction of arrival
- doa estimation
- higher order
- signal subspace
- sensor array
- canonical correlation analysis
- neural network
- feed forward
- echo state networks
- artificial neural networks
- natural images
- estimation error
- correlation matrix
- pairwise
- linear array
- sound source
- noisy environments
- background noise
- hyperspectral
- signal to noise ratio
- image segmentation
- image processing
- noisy images
- parameter estimation
- speech enhancement
- signal processing
- least squares
- multiscale
- face recognition