Noise-Separated Adaptive Feature Distillation for Robust Speech Recognition.
Honglin QuXiangdong SuYonghe WangXiang HaoGuanglai GaoPublished in: IEEE Signal Process. Lett. (2023)
Keyphrases
- speech recognition
- noisy environments
- language model
- speech signal
- spectral subtraction
- hidden markov models
- automatic speech recognition
- speech processing
- cepstral coefficients
- speech synthesis
- noisy speech
- speech recognition technology
- speech enhancement
- speaker identification
- speech recognizer
- speech recognition systems
- digit recognition
- speech understanding
- pattern recognition
- background noise
- noise reduction
- word recognition
- additive noise
- noise level
- signal to noise ratio
- feature vectors
- natural language
- keyword spotting
- noisy images
- speaker independent
- speaker dependent
- neural network