Hi! I have just started messing around with DAS to analyze mouse USVs. Working locally on a windows PC with a nvidia GPU, installed via conda and gui is working just fine. I have opened a wav file that I have bandpassed filtered between 20khz-115khz, as I have a decent amount of background noise in the audible range. I have a few examples of USVs in my wav file that I can see, but hitting E doesnt result in any sound. of course when I dont bandpass and do the same, I can hear the background noise, so I imagine it is trying to play an ultrasonic tone through my less than capable speakers.
Is there a way to downshift the USVs down into the audible spectrum for playback/monitoring purposes only (similar to the behavior of DeepSqueak)? I searched around the settings and the docs and did not find anything. I'd rather not have to create new downshifted copies of all the files.
Thanks in advance for the help! if this is not an option, it would be an awesome feature. Relying purely on the spectrogram is suboptimal, plus 90% of the fun of this work is listening to the mice sing in slow motion!
Hi! I have just started messing around with DAS to analyze mouse USVs. Working locally on a windows PC with a nvidia GPU, installed via conda and gui is working just fine. I have opened a wav file that I have bandpassed filtered between 20khz-115khz, as I have a decent amount of background noise in the audible range. I have a few examples of USVs in my wav file that I can see, but hitting E doesnt result in any sound. of course when I dont bandpass and do the same, I can hear the background noise, so I imagine it is trying to play an ultrasonic tone through my less than capable speakers.
Is there a way to downshift the USVs down into the audible spectrum for playback/monitoring purposes only (similar to the behavior of DeepSqueak)? I searched around the settings and the docs and did not find anything. I'd rather not have to create new downshifted copies of all the files.
Thanks in advance for the help! if this is not an option, it would be an awesome feature. Relying purely on the spectrogram is suboptimal, plus 90% of the fun of this work is listening to the mice sing in slow motion!
Thanks