Improving Discriminative Training for Robust Acoustic Models in Large Vocabulary Continuous Speech Recognition.
Janne PylkkönenMikko KurimoPublished in: INTERSPEECH (2012)
Keyphrases
- discriminative training
- acoustic models
- automatic speech recognition
- speech recognizer
- hidden markov models
- speech recognition
- probabilistic interpretation
- maximum likelihood
- noisy environments
- posterior probability
- broadcast news
- speech signal
- data sets
- speaker independent
- spoken language
- similarity measure
- text mining
- probability distribution
- training data