- This topic has 3 replies, 3 voices, and was last updated 7 years ago by .
Viewing 3 reply threads
Viewing 3 reply threads
- You must be logged in to reply to this topic.
› Forums › Automatic speech recognition › Dynamic Time Warping (DTW) › Summing up over ALL path costs?
True or false: “A key operation in DTW is the summation of all path costs at each point in the grid.”?
What confuses me is the term “ALL” – would that imply the addition not only of subpaths that make up ONE path leading into the cell, but of all the possible (alternative paths)? Because the latter is false.
In DTW, there is no summing at all. We are only interested in the signal best path, and so take the minimum path cost at every grid point.
Yes, but to we do sum up over sub-paths that belong to the SAME (sub)path, leading into that node? Else how do we get the total path cost? And wouldn’t that be too greedy otherwise?
Every time a token moves, you’re updating its distance with a subcost, not recomputing the distance by summing up all past subcosts. The question’s referring to what happens at a junction with regard to all incoming tokens.
Some forums are only available if you are logged in. Searching will only return results from those forums if you log in.
Copyright © 2024 · Balance Child Theme on Genesis Framework · WordPress · Log in