Speech Enhancement Based on Teacher-Student Deep Learning Using Improved Speech Presence Probability for Noise-Robust Speech Recognition.
Yanhui TuJun DuChin-Hui LeePublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2019)
Keyphrases
- speech recognition
- noisy environments
- speech enhancement
- deep learning
- speech signal
- noisy speech
- spectral subtraction
- automatic speech recognition
- background noise
- speaker identification
- hidden markov models
- language model
- speech recognizer
- noise reduction
- pattern recognition
- speech synthesis
- machine learning
- additive noise
- signal to noise ratio
- speech recognition systems
- vocal tract
- mental models
- unsupervised learning
- learning environment
- image processing
- feature selection
- learning algorithm