Leveraging threshold denoising on DCT-based modulation spectrum for noise robust speech recognition.
Yen-chih ChengJun-Shan LinJeih-weih HungPublished in: ICCA (2014)
Keyphrases
- speech recognition
- noisy environments
- spectral subtraction
- denoising
- speech signal
- hidden markov models
- automatic speech recognition
- gaussian noise
- signal to noise ratio
- noise removal
- noisy speech
- language model
- speech synthesis
- speech processing
- speaker identification
- background noise
- image compression
- speech recognizer
- speech enhancement
- speech recognition technology
- speech recognizers
- image denoising
- denoising methods
- noise reduction
- speech recognition systems
- pattern recognition
- speech retrieval
- natural images
- wavelet domain
- neural network
- power spectrum
- noisy images
- image restoration
- image processing