On Maximizing the Within-Cluster Homogeneity of Speaker Voice Characteristics For Speech Utterance Clustering.
Wei-Ho TsaiHsin-Min WangPublished in: ICASSP (1) (2006)
Keyphrases
- speech recognition
- clustering algorithm
- speech synthesis
- data clustering
- hierarchical clustering
- automatic speech recognition
- speech signal
- cluster analysis
- audio visual
- speaker recognition
- speaker identification
- speech sounds
- prosodic features
- speech recognition errors
- data points
- synthesized speech
- text to speech
- clustering framework
- cluster centers
- inter cluster
- clustering method
- spoken language
- clustering scheme
- speech recognizer
- intra cluster
- clustering approaches
- hidden markov models
- emotion recognition
- validity measures
- disjoint clusters
- voice activity detection
- overlapping clusters
- k means
- clustering procedure
- speech quality
- vocal tract
- mel frequency cepstral coefficients
- speaker diarization
- language model
- agglomerative hierarchical clustering
- pattern recognition
- arbitrary shape
- hierarchical clustering algorithm
- clustering result
- constrained clustering
- fuzzy clustering
- cluster membership
- clustering quality
- similar objects
- cluster validation
- clustering solutions
- unsupervised clustering
- homogeneous groups
- speaker verification
- unsupervised learning
- clustering analysis
- fundamental frequency
- speaker independent
- noisy environments
- density based clustering algorithm
- speaker dependent
- hierarchical agglomerative clustering
- spectral clustering
- model based clustering
- self organizing maps