Ideal Ratio Mask Estimation Using Deep Neural Networks for Monaural Speech Segregation in Noisy Reverberant Conditions.
Xu LiJunfeng LiYonghong YanPublished in: INTERSPEECH (2017)
Keyphrases
- neural network
- speech signal
- noisy environments
- speech recognition
- sufficient conditions
- automatic speech recognition
- pattern recognition
- artificial neural networks
- vocal tract
- hidden markov models
- parameter estimation
- neural network model
- accurate estimation
- estimation accuracy
- speaker recognition
- standard deviation
- text to speech
- multilayer perceptron
- feed forward
- fuzzy logic
- motion estimation
- genetic algorithm
- speaker identification
- recognition engine
- robust estimation
- multi layer
- estimation error
- neural nets
- recurrent neural networks
- noisy data
- self organizing maps