Mean temporal distance: Predicting ASR error from temporal properties of speech signal.
Hynek HermanskyEhsan VarianiVijayaditya PeddintiPublished in: ICASSP (2013)
Keyphrases
- temporal properties
- speech signal
- automatic speech recognition
- speech recognition
- model checking
- noisy environments
- speaker identification
- vocal tract
- background noise
- linear prediction
- possibility theory
- hidden markov models
- spectral analysis
- automatic speech recognition systems
- non stationary
- formal methods
- broadcast news
- speech enhancement
- speech quality
- speaker recognition
- fundamental frequency
- machine learning
- incomplete information
- pattern recognition
- image processing