| 摘要 |
Invention refers to a system that enables a detection of positive psycho emotive condition of a speaker on the grounds of analysis of the voice, quantification and classification of that condition. System is accomplished in the frames of speaking and telecommunicating technology (fixed and mobile telephony). The system consists from an input (1) and output module (2) which enables interface with telecommunicating medium (6). Processing a speech signal is carried out on a software platform (2) on the grounds of a group of software modules (3) and a psycho emotive module base (4). Upon 3D treatment of a speech signal (7) carries out extraction of marks (8, 10) based on accoustic and emotional modules (9), (11) and thereupon carries out integration of information on the base of psychological models (12), (13). Finally, recognition of speech expression of emotions (14) is accomplished by means of a psycho emotive module base (15).
|
