Comparing one and two-stage acoustic modeling in the recognition of emotion in speech

Keyphrases

• Acoustic Modeling 100%
• Acoustic Properties 50%
• Beam Search 50%
• Brute Force 50%
• Chunk Analysis 100%
• Feature Space 50%
• Full Section 50%
• High Efficiency 50%
• High Temporal Resolution 50%
• Multiple Instance Learning 50%
• Pre-segmentation 50%
• Property Instances 50%
• Quasi-stationary 50%
• Recognition of Emotions in Speech 100%
• Space Construction 50%
• Speaker Normalization 50%
• Stage Acoustics 100%
• Subset Selection 50%
• SVM Classification 50%
• Token Passing 50%
• Two-stage Approach 50%
• Viterbi 50%

Computer Science

• Feature Space 100%
• Feature Subset Selection 100%
• Instance Learning 100%
• Presegmentation 100%
• Processing Stage 100%
• Support Vector Machine 100%
• Temporal Resolution 100%