1 / 3

Particle filter – IFC implementation:

X(k) = A * X(k-1) + V(k) Y(k) = B * X(k) + U(k). Where, A transition matrix B observation matrix. Accept file (one frame at a time). Initial processing**. Assign values to parameters**. Generate particles. observation. “states estimates”. Predict states. Update states.

medea
Download Presentation

Particle filter – IFC implementation:

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. X(k) = A * X(k-1) + V(k) Y(k) = B * X(k) + U(k) Where, A transition matrix B observation matrix Accept file (one frame at a time) Initial processing** Assign values to parameters** Generate particles observation “states estimates” Predict states Update states Importance weights Predict states More observations?? • Compute autocorrelations, LPCs, noise variances. • A and B • Based on std devn and initial obsns • Based on A, V(k) • Based on predicted states, B, current obsn, • Resampling of states • results • continue • Particle filter – IFC implementation:

  2. Results from Kalman and Particle filtering IFC Original signal Kalman-filtered particle-filtered

  3. Particle Filtering for filtering: • Kalman filtering implementation gives some reasonable results: • Order affects the results. • Tracking (/ filtering) of the speech is possible. • Particle filtering as used for filtering (similar to Kalman filter implementation) : • Computation results at each block are correct (mathematically). • Has different results (strange results) as compared to the one had with Kalman filter. • Reasons: (probable) • Lesser number of particles • Lesser order value • Noisy signal cannot be modeled by a single Gaussian distribution. • Modeling of speech signal in the way done is flawed. • Code has some serious problems [Huh?] Ruled out. (different number of particles tried) 50, 100, 700 Ruled out. (different orders tried) 5, 8, 10 Ruled out. But code has not yet been reviewed.

More Related