Reliable accent specific unit generation with dynamic Gaussian mixture selection for multi-accent speech recognition.
Chao ZhangYi LiuYunqing XiaThomas Fang ZhengJesper Ø. OlsenJilei TianPublished in: ICME (2011)
Keyphrases
- speech recognition
- gaussian mixture
- hidden markov models
- automatic speech recognition
- mixtures of gaussians
- speech recognizer
- speech signal
- speech recognition systems
- speech recognition technology
- language model
- speech synthesis
- speech processing
- mixture model
- covariance matrix
- closed form
- em algorithm
- expectation maximization
- pattern recognition
- noisy environments
- speaker identification
- probability density function
- isolated word
- data mining
- high resolution
- objective function
- similarity measure