Enhancing the intelligibility of statistically generated synthetic speech by means of noise-independent modifications.
Daniel ErroTudor-Catalin ZorilaYannis StylianouPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2014)
Keyphrases
- signal to noise ratio
- speech recognition
- noisy environments
- noisy data
- noise level
- audio visual
- speech signal
- random noise
- automatically generated
- additive noise
- randomly selected
- automatic speech recognition
- audio stream
- speech enhancement
- recognition engine
- image noise
- noise removal
- data sets
- gaussian noise
- language model