Disabled Assisting Intelligent On Speech Recognition And Control System Based On DSP

Speech recognition is a kind of comprehensive technology which has relationship with acoustics, linguistics, digital signal processing, etc., because of its outstanding human-machine interface, it is becoming a research hotpot in the information technology as an important research field of the series of handicapped-aid robots. Disabled people are vulnerable groups in society, through the application of high-tech assistive service robots, they can overcome physical disability, and do what a normal person can do, self-reliant. So it is a manifestation of social progress and development. Therefore, researching practical intelligent wheelchair speech recognition controlling system not only has a good meaning of theory but also has a great value in practicality.Based on all above these, this paper studies intelligence speech recognition system which takes TMS320VC5402 as the core. The system takes the TMS320VC5402 chip as a core, and carries on TMS320VC5402 the exterior memory's expansion. The system carries on the mold/number by TLC320AD50C to transform, carries on the voice signal training and the recognition by TMS320VC5402. System's logical interface is realizes by logic chip CPLD. Considered the disabled assisting device,the intelligent wheelchair's characteristic take non-specific person,particular speech (like "qian"), LPCC endpoint detection system applications for voice signal parameters and to establish reference model database, using dynamic time warping algorithm for treatment of the voice signal matching recognition test.