Improving Automatic Speech Recognition Containing Additive Noise Using Deep Denoising Autoencoders of LSTM Networks.
Marvin Coto-JiménezJohn Goddard CloseFabiola Martínez LiconaPublished in: SPECOM (2016)
Keyphrases
- denoising
- additive noise
- noisy images
- automatic speech recognition
- speech signal
- image denoising
- speech recognition
- noise level
- total variation
- conversational speech
- causal inference
- natural images
- word error rate
- hidden markov models
- image processing
- broadcast news
- denoising algorithm
- wavelet domain
- spoken words
- wiener filter
- gaussian noise
- frequency domain
- noisy environments
- denoising methods
- random variables
- wavelet packet
- speech retrieval
- recognition errors
- noise model
- word recognition
- image segmentation
- vocal tract
- recurrent neural networks
- non stationary
- linear prediction
- bayesian networks
- signal processing
- pattern recognition
- computer vision
- graphical models
- language model