Question about sonogramming my recordings...

[Steelflight]

Ok... so I have a recording of a great horned owl, and it is probably a couple hundred feet away. The ME66 definitely picked it up, because I can hear the recording. However, when I try to sonogram it, the call doesn't stand out at all against the other noise. Is there anyway to "bring out" the faint calling on my recording? Or is my only option moving in a lot closer?

Thanks again : ]

[Steelflight]

Also, how do I post pictures of my sonograms on these forums? I don't know how to get it in the right format so I can attach it to a message.

[ermine]

However, when I try to sonogram it, the call doesn't stand out at all against the other noise. Is there anyway to "bring out" the faint calling on my recording? Or is my only option moving in a lot closer?


The computer between your ears is a lot more powerful than the one in front of you, so on a really marginal recording you may find you can't get a sonogram even if you can hear the call. I don't claim to be a sonogram expert but -

have you normalised the signal - ie raised the gain so the owl is full-scale? The background noise will rise up massively too, but I find Raven doesn't perform well unless the signal occupies the full gain range. Naturally do this with a copy to feed to the program rather than changing you original.

Filtering and noise reduction can make the sound clearer, but are frowned on for sonograms because you don't know whether it is the signal or the processing you are making a sonogram of.

Getting closer is always good, and by far the best way.

As for image formats, they are on the attach files option in additional options when you post.

Valid file extensions: bmp doc gif jpe jpeg jpg mpeg pdf png psd txt wmv

[Steelflight]

Ok so how do I make the sonogram a jpeg or gif "picture"? When I save my Audacity or Raven sonograms, they are sound files... no?

[ermine]

Ok so how do I make the sonogram a jpeg or gif "picture"? When I save my Audacity or Raven sonograms, they are sound files... no?

File -> Export Image of -> Raven Application Window

(or the particular window you are interested in). You get better results if you adjust the windows in Raven to be the right size you want the image to be rather than rescaling afterwards.

Exporting the app window at least shows you the settings for the currently highlighted window. Raven doesn't appear able to save these sessions, or maybe that's what you get when you stump up the extra for the man's version rather than $0 the Lite version.

There is much variation is the visibility of the signal depending on the Raven settings. F'rinstance, the attached recording of a chiffchaff (http://www.rspb.org.uk/wildlife/birdguide/name/c/chiffchaff/index.asp) sounds reasonably clear to me. And yet the default Raven settings show a lot of murk (image right hand side) and adjusting them brings out the pattern that gives this bird it's curious tone colour. You can see the settings used in the capture which are a long way away from the defaults.

[Steelflight]

Hey thanks emine!

Is there a way to do it in Audacity (that is, export a copy of the sonogram)? For some reason my sonogram (waveform) is showing up a lot clearer in Audacity than it is in Raven. Thanks

Or, do you have a suggestion for the best program to use for printing out sonograms? Thanks.

[ermine]

[QUOTE=Steelflight]Hey thanks emine!

Is there a way to do it in Audacity (that is, export a copy of the sonogram)? For some reason my sonogram (waveform) is showing up a lot clearer in Audacity than it is in Raven. Thanks

QUOTE]

Press the PrintScreen button to take a screenshot of your pc screen showing desired sonogram in Audacity. Then paste as new document in your preferred drawing packing, and crop to take off all the spurious desktop chrome.

If Audacity shows it well then Raven can. You'll be better off in the long run learning how to control Raven to give you what you want because you can control the settings, with Audacity you just take it as it comes and have no idea of whether the settings are repeatable. The contrast and brightness controls should let you pull most stuff out of the murk. Increase contrast a bit, adjust brightness to give the right balance between clarity and noise, adjust contrast a bit more and go round the loop a couple of times. The third control with the three bars adjusts the balance between frequency resolution (vertical scale) and temporal resolution (horizontal scale). If you're looking to see the harmonic structure of a finch call, for instance, you want to be higher than the 256 default.

This also depends on how much you zoom in to show a particular phrase - Raven's settings are about right if you zoom in to show only 1s. If you view a larger segment then Raven is set to show a bit too much temporal resolution which you can't see because the details are less than a pixel wide, whereas increasing frequency resolution is visible at both wide and narrow time spans. For your owl hoots try and push that value up somewhat - I'm guessing but there's probably not much fine temporal detail in an owl hoot compared to a songbird, and increasing frequency resolution can help pick out the visual pattern agains the background noise until the sonogram starts to break up.

[Steelflight]

Hey, thanks ermine-- starting to like Raven more and more!

Ok, now that I know how to make pictures of my sonograms-- I still have one problem.

For some reason Raven won't let me "filter" or "amplify" my sonograms or spectrograms. I highlighted a section, and then for some reason the "amplify" and "filter" buttons are gray-- I can't click them! I checked the manual, and it says to just highlight a selection, then click the button.

How come the buttons are grayed out? I have a selection highlighted... but for some reason it won't let me do either of those actions.

[Steelflight]

For some reason it lets me do it on the "Example sounds" that come with Raven, but it won't let me amplify / filter my sound files.

[ermine]

For some reason it lets me do it on the "Example sounds" that come with Raven, but it won't let me amplify / filter my sound files.

Doesn't let you do that on a stereo recording, because Raven only thinks in mono. It doesn't know what you want to do with the other channel. Maybe the man's version of Raven does that - there has to be some reason for putting down the $$$, but your recordings are mono anyway. So convert to mono first. While you're at it, do your amplifying normalization and filtering in Audacity before you convert to mono, as Raven's capabilities for doing that aren't so great anyway.

[Steelflight]

Thanks emine :).

Got another question for you--> is there a way to find the maximum frequency of a particular segment of a bird's call, other than just using the 'trace' feature and going to the highest part of a spectrogram?

Thanks