Robust speaker identification using combined feature selection and missing data recognition.
Daniel PullellaMarco KühneRoberto TogneriPublished in: ICASSP (2008)
Keyphrases
- missing data
- speaker identification
- noisy environments
- feature selection
- feature extraction
- missing values
- incomplete data
- speech recognition
- matrix factorization
- automatic relevance determination
- machine learning
- low rank
- speech signal
- noise reduction
- perfect phylogeny
- multiple imputation
- pattern recognition
- feature space
- hidden markov models
- text categorization
- broadcast news
- text classification
- feature set