Sparse coding of the modulation spectrum for noise-robust automatic speech recognition.
Sara AhmadiSeyed AhadiBert CranenLou BovesPublished in: EURASIP J. Audio Speech Music. Process. (2014)
Keyphrases
- sparse coding
- automatic speech recognition
- noisy environments
- dictionary learning
- speech recognition
- sparse representation
- natural images
- image classification
- linear combination
- speech signal
- norm minimization
- unsupervised learning
- hidden markov models
- image patches
- noise reduction
- image representation
- generative model
- sparse codes
- conversational speech
- signal to noise ratio
- visual words
- noise model
- object recognition
- high dimensional
- non stationary
- feature space