Summary: | 碩士 === 大同大學 === 資訊工程學系(所) === 94 === Language is a mean of communication and speech plays an important role in the society. So how to express emotion correctly is an important aspect in communication. In this thesis, emotion recognition from continuous Mandarin speech signal is implemented. In the experiment, Mel-Frequency Cepstral Coefficients (MFCC) and Linear Prediction Cepstral Coefficients (LPCC) are selected as the features used in the recognition. Five emotions are investigated, including anger, happiness, sadness, boredom, and neutral.
Endpoint detection is tried to segment the continuous speech. Weighted discrete K-nearest neighbor method is chosen as the classifier. In this research, the continuous sentences are composed by several short sentences with known emotion. The average recognition accuracy is 83% for these sentences.
|