Posterior-based sparse representation for automatic speech recognition.
Sara BahaadiniAfsaneh AsaeiDavid ImsengHervé BourlardPublished in: INTERSPEECH (2014)
Keyphrases
- sparse representation
- automatic speech recognition
- speech recognition
- dictionary learning
- sparse coding
- speech signal
- word error rate
- hidden markov models
- compressive sensing
- conversational speech
- image classification
- face recognition
- image representation
- signal processing
- image patches
- dimensionality reduction
- broadcast news
- joint optimization
- compressed sensing
- regularized least squares
- probabilistic model
- bayesian framework
- random projections
- test images
- high dimensional data
- speech corpus
- speech retrieval
- gaussian process
- principal component analysis
- high dimensional
- pattern recognition