Spoken English Intelligibility Remediation with PocketSphinx Alignment and Feature Extraction Improves Substantially over the State of the Art.
Yuan GaoBrij Mohan Lal SrivastavaJames SalsmanPublished in: CoRR (2017)
Keyphrases
- feature extraction
- speech recognition
- spoken language
- broadcast news
- word level
- word alignment
- preprocessing
- english language
- pattern recognition
- feature vectors
- natural language
- speaker identification
- feature extractors
- frequency domain
- principal component analysis
- machine translation
- texture features
- discriminant analysis
- wavelet transform
- language learning
- extracted features
- feature extraction and classification
- cross lingual
- pattern classification
- face recognition
- image preprocessing
- cross language
- signal to noise ratio
- feature representation
- automatic speech recognition
- english text
- image alignment
- statistical machine translation
- linear discriminant analysis
- feature space
- speech retrieval
- sentence pairs
- feature selection
- language understanding
- language processing
- cross language information retrieval
- palmprint
- support vector machine svm
- image classification
- feature set