Topic Speech enhancement

profileshuhuan123

http://en.wikipedia.org/wiki/Wiener_filter

Based on the link and the attachment PDF file, write a 10pages essay, the framework of the essay should be like the following :

 

 Topic  Speech enhancement

① Back ground

②Utilization

③How it works

Characters of speech signal(random, change overtime, unstable)

Ear → machine   Cover effect---how human ear can distinguish speech signal in noisy environment

Interference: (noise stable interference)  --characteristics 

Ways of speech enhancement

  1 Based on time domain

     a.Linear prediction analysis---estimate parameter of model --combining clean speech signal

     b.Only use the parameter of sound 

Wiener filtering ----replenish----Kalman filtering   eta(t)=s(t)+n(t) (mean-square deviation least ) e(n)=s(n)-s’(n)→E[e2(n)]=E[S-S’2]

Based on all data in the past and the current data by watch---estimate current value

 

6.Kalman filtering

  • 11 years ago
  • 20
Answer(1)

Purchase the answer to view it

blurred-text
NOT RATED
  • attachment
    speech_enhancement-_final_copy.docx