I have ultrasonic vocalisation recordings from rats that I would like to analyse, though I am finding it difficult what methods and programs are best to analyse them? In particular I am interested in free software that show spectrograms, allows playback in the audible range and can delivery some frequency characteristics.
Hi Nathanael,
at a first glance try audacity. It is free and has all You want (spectogram, FFT). You can also change a sample rate to make the sound audible.
Good luck !
Chris
I've found that seewave for R is a good tool, especially when used with R's native locator function (so you can select start and stop times in the graphics window). I built a simple function to auto-open a cut with different resolution (one for high temporal resolution and a second with high frequency resolution), thus allowing me to fly through sound files. It also has powerful tools for analysis. And, of course, you have direct access to R itself once you get into the post-processing and analysis phases. Perhaps use Audacity to isolate cuts and then move your workflow to R for analysis of the cuts.
Thanks for your comments. I ended out developing my own program using Labview. If anyone is interested in using it please let me know and I will pass it on to you.
Are you able to distinguish the vocalisation of the rats or can you find matches (recognize rats) in your ultrasonic records?
Hi Marc, I haven't tried this out. I have only performed it on rat pups ~PND10. It may be possible with sufficient analysis with machine learning - but would probably be difficult practically. There is also the issue that the calls are very directional, meaning that you are unlikely to pick up calls of equal intensity from multiple animals at the same time.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping