Combining dynamic texture and structural features for speaker identification.
Guoying ZhaoXiaohua HuangYulia GizatdinovaMatti PietikäinenPublished in: MiFor@MM (2010)
Keyphrases
- structural features
- speaker identification
- dynamic textures
- speech recognition
- noisy environments
- gaussian mixture model
- speech signal
- structural information
- broadcast news
- feature set
- feature extraction
- spatial and temporal
- semantic features
- space time
- optical flow
- generative model
- mixture model
- natural language
- pattern recognition
- distance measure
- spatio temporal
- training data