A new data-driven band-weighting function for predicting the intelligibility of noise-suppressed speech.
Zexin LiuHeather T. MaFei ChenPublished in: APSIPA (2017)
Keyphrases
- data driven
- speech recognition
- signal to noise ratio
- noisy environments
- speech enhancement
- noise level
- noise reduction
- speech signal
- noisy data
- random noise
- noise model
- low frequency
- pattern recognition
- automatic speech recognition
- image structure
- neural network
- low signal to noise ratio
- speaker identification
- spoken language
- background noise
- missing data
- additive noise
- median filter
- audio visual
- low pass
- multiscale
- high frequency
- data sets
- speech synthesis
- noise sensitivity
- recognition engine
- language model