End-to-end recurrent denoising autoencoder embeddings for speaker identification.
Esther Rituerto-GonzálezCarmen Peláez-MorenoPublished in: Neural Comput. Appl. (2021)
Keyphrases
- end to end
- speaker identification
- denoising
- speech recognition
- image denoising
- gaussian mixture model
- speech signal
- noisy images
- noisy environments
- broadcast news
- image processing
- natural images
- congestion control
- feature extraction
- low dimensional
- dimensionality reduction
- high dimensional data
- real time
- image data
- feature space
- image sequences