search

TWOEARS / auditory-front-end

Two!Ears Auditory Model - Auditory front-end module
http://docs.twoears.eu/en/latest/afe/
GNU General Public License v2.0
16 stars 10 forks source link

Ensure processors return values that are independent of sampling frequency #5

Closed hagenw closed 7 years ago

hagenw commented 8 years ago

This might be the easiest solution to avoid problems due to the usage of different sampling frequencies as described in TWOEARS/blackboard-system#10.

For example for the cross correlation processor we could return the lag in seconds and not in samples.

hagenw commented 8 years ago

Is there an update on this? In our real live example we will have audio with 44.1 kHz and would like to test the DNNLocationKS which depends on the cross correlation processor. @Hardcorehobel @ningma97 could you please change the behavior of the cross-correlation processor to return the lag values in seconds and adjust DNNLocationKS accordingly? @Hardcorehobel: if you don't have time please indicate this too, then we will have a look at this here in Toulouse. /cc @chrschy

hagenw commented 7 years ago

We found a intermediate solution for this, so I will close this. If new problems in this direction occur, we could open a new issue.