Regularized subspace Gaussian mixture models for cross-lingual speech recognition.
Liang LuArnab GhoshalSteve RenalsPublished in: ASRU (2011)
Keyphrases
- cross lingual
- speech recognition
- gaussian mixture model
- language model
- language modeling
- mixture model
- speaker identification
- speaker recognition
- feature space
- machine translation
- language independent
- automatic speech recognition
- feature vectors
- probabilistic model
- cross language
- principal component analysis
- hidden markov models
- pattern recognition
- text classification
- mel frequency cepstral coefficients
- speech signal
- noisy environments
- maximum likelihood
- em algorithm
- expectation maximization
- query expansion
- low dimensional
- dimensionality reduction
- information retrieval
- computer vision
- news articles
- high dimensional
- retrieval model
- n gram
- document clustering
- language identification
- parallel corpora
- feature extraction