• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
ZEISS DTI thermal imaging cameras. For more discoveries at night, and during the day.

searching spectrograms (1 Viewer)

craigwilson

Well-known member
New Zealand
Given an extended (2+ hour) dawn chorus recording and a known sonogram/spectrum specific to a certain bird, is there a way to search and find in the extended recording each instance of this bird's vocalisations using Audacity or Raven Lite?
 
I don't think so. Raven Pro lets you search, but not by the actual shape of the call on the spectrogram. You can only specify the frequency range and duration of what you're looking for, and I think you can also specify some parameters about the time between calls. It marks them all, and then you check them manually.

It's very fiddly, and easy to set it to mark too many or too few. I tried it on boobook calls, with mixed success.

If you're only doing one recording, it's probably not so hard to scan through the recording manually in Audacity using Page Up. I believe in the latest version of Audacity you can mark regions on the spectrogram with a frequency range as well as a time range on a label track, which might be useful.

I haven't tried Raven Lite for a long time. I think it used to limit the length of the track you could examine, but I think that might have been removed in the latest version. If it allows marking of time/frequency ranges like the Pro version, it's much more suited to marking calls than Audacity, but if it's anything like the Pro version was last time I looked, it's lacking in shortcut keys to make scanning through the track simple.

I'm really interested to hear what you end up using.
 
Thanks for your message. Raven Lite still puts a limit on recording length I believe. Here's a link to some sound analysis software 'not for the faint of heart' suggested by a guy at the Audacity forum.
http://songbirdscience.com/resources/behavior/sound-analysis-software
Luscinia seems the most promising, several are windows only - will need to carve out a chunk of time to tackle this.

I have found stationary 2hr + recording most interesting as species are out there I have not yet seen/identified.
 
Luscinia does sound promising, but very complicated. I vaguely remember playing with Praat, but not getting anywhere. I think I found it too oriented towards speech analysis.

I've found the same thing with long recordings. Species that were there but I never heard or saw them myself at the time, like buttonquail and Malleefowl.

Please let us know how you get on.

PS Raven Lite is probably unsuitable anyway, but for the record, it is now time unlimited. Here are the specs for the latest version:
* Sound view window

o No limit on sound length

o Paged sound windows

o Multi-channel sound views

o Waveform, spectrogram, spectrogram slice and selection spectrum views

o Real date and time display

* Selection tables - multiple selections for

o zoom

o playback

o annotation

o "Average Power Density (dB FS)" measurement

o export to sound clips

* Customization

o User-defined presets for sound windows, recorders, and color schemes

o Raven Preferences

* Recorder

o No limit on recording time

o Reads multi-channel audio into memory, single files, or file sequences

o Multiple sample rates

o Decimation of audio samples

o Delayed recording

o File naming templates

* Playback

o Scrolling playback

o Reverse playback

o Looping playback

* Filters

o Bandpass and bandstop

o Adaptive

o Amplify

o Fade in and fade out

* Memory Manager

o user can set Raven Lite to use more memory for longer sound views and higher sample rates
 
Turns out there are heaps of published papers addressing this problem. I do not have the mathematic and acoustic training to understand most. There is the proprietary Song Scope ($$) however - the methods for ID in 876 hrs of recording are well outlined in this paper: 1. Buxton RT, Jones IL. Measuring nocturnal seabird activity and status using acoustic recording devices: Applications for island restoration. J F Ornithol. 2012;83(1):47–60. And here's an evaluation: 1. Wolfgang A, Haines A. Testing Automated Call-Recognition Software for Winter Bird Vocalizations. Bioone. 2016;23(2):249–58.
 
Warning! This thread is more than 7 years ago old.
It's likely that no further discussion is required, in which case we recommend starting a new thread. If however you feel your response is required you can still do so.

Users who are viewing this thread

Back
Top