birding-aus

Sound Recording Bird Calls

To: Peter Shute <>
Subject: Sound Recording Bird Calls
From: Paul McDonald <>
Date: Mon, 22 Aug 2011 12:18:11 +1000
Dear Peter, 

This is tough one to answer, as it depends upon what you want to use the sounds 
for. If they are to be used as a guide for birders to learn species X call in 
Location Y, than mp3 is fine. If they are to look at syllables (subunit of a 
song) in terms of presence absence, than mp3 is 'probably' OK for manual 
scoring by ear. For example, a species might sing a song that goes 'Tweet chirp 
trill'. Mp3 is probably good enough quality to see if in a different population 
it sings these syllables (tweet, chirp, trill). They may, for example, sing 
instead 'Trill burp tweet', thus introducing a new syllable,  losing another 
and changing the order that syllables are presented. This might all be doable 
with mp3 files if the song is fairly simple as above. More complex calls/songs 
would need statistical assessment, and that's where mp3 won't cut it. 

Further, if you then wanted to try and identify individuals from the calls, 
e.g. as they have done in the past with Corn crakes, Sacred Kingfishers and so 
on, you need better quality audio. Also, if you wanted to look at regional 
differences in the calls (ie bandwidth, relative amplitudes/power at various 
frequencies and so on that might be indicative of things such as dialects) than 
the mp3 is not suitable - you need PCM quality for these types of analyses. 
You'd also need that level of detail to prove subspecies, such as in the 
Cicadabird example. 

The problem is that the calls are changed when converted or recorded into mp3 
format in a relatively unpredictable way, and this means that we cannot 
recreate or even identify what has been lost. 

Sound is pretty low in terms of memory requirements, my picture/video folders 
are much larger than those I have for sound. Memory is becoming cheaper all the 
time (store things on hard drives not DVD/CDs that may degrade), I use Lacie 
Rugged drives that have proven very reliable, 1TB drive is now just $250 and 
could be used to archive any file, not just sound. YOu can also cut out the 
bits of interest, and remove the silence between songs, as well as strip out a 
channel that isn't relevant, ie no point recording in stereo if you have only 
one mic, so dropping the file back to mono saves almost half of your file size 
and so on.

I haven't used the Sony, but remember reading a positive review somewhere or 
other, and your solution of PVC pipe as a shockmount is likely as effective as 
the expensive Rycote grips. 

Haven't recorded on DSLRs, sadly don't own one. The caveats outlined below 
would all still apply in terms of the quality that they could provide. 
Connecting a shotgun mic would improve things markedly I'd imagine, and might 
simplify kit in the field. However, I haven't looked into them. 

I guess in short, something would be better than nothing (ie we'd love a call 
from the dodo, regardless of quality!), but if you can record in a better 
format that is suitable for any application that is likely to arise. In this 
way we have the most information going forward. If serious about it, you should 
also invest in good quality gear. Each of the bottlenecks of sound recording 
revolve around key pieces: a bad microphone will not record a song well, 
regardless of how good the recorder it is attached to is and so on.  

I hope that this helps, 
Paul

 

 

 


On 22/08/2011, at 10:13 AM, Peter Shute wrote:

> Paul, can you please elaborate on the issue of mp3 format vs PCM format? I 
> initially used PCM after reading advice similar to yours, but found the file 
> sizes involved daunting in terms of storage and download times, not to 
> mention backup times, so I reverted to mp3 format.
> 
> As you said, they both sound the same to me. Are they totally useless for 
> analysis, or just less useful? If the former then people considering buying 
> equipment might want to factor the cost of extra memory cards and hard disks 
> into it.
> 
> I'd also like to make a comment about choosing equipment that I would hope 
> might be discussed at the WA workshop. One problem I had was local 
> availability. Perhaps I was trying the wrong shops, but I could only find the 
> occasional camera shop with a small range of Olympus recorders at exhorbitant 
> prices (well over double the internet prices), and music shops refuse to sell 
> anything but Zooms (but at good prices).
> 
> I didn't want a Zoom, despite them being multi track and having phantom 
> power, because I was told they are slightly noisier than Olympus and have far 
> less battery life. I wanted to try an Olympus LS-7, which apparently is 
> quieter than the LS-11 and has a pre record buffer, but wan't going to risk 
> buying sight unseen something very few nature recordists were using.
> 
> I ended up with a Sony PCM-M10, which appears to be well regarded and widely 
> used despite not being mentioned in this thread yet. Like many of the other 
> models that have been mentioned here, it was only about $250 from for B&H in 
> the USA. I simply couldn't find one here.
> 
> I bought a Sennheiser K6/ME66 shotgun microphone on eBay for $300 with a wind 
> cover, and made my own mount from PVC pipe and rubber bands. So I believe 
> it's possible to get set up for way less than the figures quoted previously, 
> and it's all smaller and lighter than a DSLR and lens.
> 
> I'm also interested to hear what the experts think about recording on DSLRs. 
> Many of these now do video, and have an external microphone jack. I see many 
> references on the net to people mounting a shotgun microphone on the camera. 
> Is this to be encouraged, or is the quality insufficient for this analysis 
> you refer to?
> 
> Peter Shute 
> 
>> -----Original Message-----
>> From:  
>>  On Behalf Of 
>> Paul McDonald
>> Sent: Monday, 22 August 2011 8:50 AM
>> To: Birding-Aus Birding-Aus
>> Subject: Re: [Birding-Aus] Sound Recording Bird Calls
>> 
>> Dear all, 
>> 
>> It would be fantastic if more people were collecting bird 
>> calls and depositing good quality ones on pages such as these 
>> discussed. 
>> 
>> It has been covered before, but I just want to point out that 
>> if you are to record calls for distribution to others, you 
>> need to record in PCM (ie uncompressed) format. If you choose 
>> a recorder that only allows compressed formats such as mp3, 
>> for example, this creates very small files, but does so at 
>> the expense of fidelity to the original sound. It is a 
>> complex process, but essentially mp3 and other compressed 
>> formats use algorithms that 'chop out' bits of sound that 
>> human ears cannot resolve (ie if two sounds are so close 
>> together that the human ear can't resolve them, but hears it 
>> as 'one' sound, the algorithm will essentially delete one). 
>> This saves space, but in terms of archiving sounds for later 
>> analysis, this is not good news. The calls are changed in a 
>> way that is impossible to revert. They sound fine to us, but 
>> to a birds ear probably very different, and the things that 
>> we cut out may well be important features that are used to 
>> define species, for example. It might also make no difference 
>> at all, but the point is we can't tell. 
>> 
>> So, if you simply want to record and listen to some sounds, 
>> than small hand held recorders are fine. However, they won't 
>> yield calls that can be examined in the way that Murray 
>> outlines below. 
>> 
>> If you can afford it, Sennheiser microphones are great (I use 
>> a ME67 with K6, but smaller mics would do, particularly for 
>> loud passerines), but as a rule any mic plugged into a unit 
>> will yield superior quality than the on-board tiny mics. 
>> Someone asked earlier about omni versus shotgun mics, if you 
>> are using an omni-directional mic you'll probably need 
>> something like a parabolic reflector to 'focus' the 
>> microphone onto what you're trying to record. These can be 
>> cumbersome in the field, so I prefer to use shot guns with a 
>> simple foam wind block over the top. Think of the difference 
>> as a bit like filed of view, omni records over a very wide 
>> arc, whereas shotgun mic record from a much smaller arc and 
>> are basically directional towards where they are pointed. 
>> 
>> Lots of different options for recorders, I use Sound Devices 
>> (beautiful but very very expensive) and Marantz PMD661 for 
>> hand held things. Note that there are smaller, cheaper 
>> models, such as Microtraks, available, but the latter has 
>> always introduced artefacts into the recording for the units 
>> I've tried, so wouldn't recommend them. Edirol and Fostex are 
>> also well known and trusted field recording brands in 
>> addition to Marantz. Lots of suppliers out there, Wingfield 
>> audio in the states, Video Guys in Australia, shop around. 
>> 
>> Recording gear is a bit like binoculars, everyone has a 
>> favourite. Some standard requirements for mine regardless of 
>> brand: Record in PCM,with sampling rates of 44.1kHz or 48kHz, 
>> 16 or 24 bit accuracy). Record some notes on weather, 
>> behaviour and putative individual identity, and you'll be 
>> gathering calls that sound great, but can also be used by 
>> anyone down the track.
>> 
>> There are general sites on what is required and so forth, so 
>> I'd also recommend becoming familiar with what is around 
>> before buying if you are considering this, but it can be a 
>> whole lot of fun... 
>> 
>> Cheers, 
>> Paul
>> 
>> 
>> 
>> 
>> 
>> On 22/08/2011, at 8:26 AM, Murray Lord wrote:
>> 
>>> A question was asked during last week's discussion on sound 
>> recording about why 
>>> birders should make more effort recording calls.  Here are 
>> two examples that come 
>>> to my mind of where more call recordings could impact on 
>> our knowledge of 
>>> Australian bird taxonomy.
>>> 
>>> First, the lurida race of boobook.  At least one recent 
>> publications [Owls of the 
>>> World 2nd edition by Koenig and Weick] treats this as a 
>> separate species.  Yet the 
>>> fieldwork that could establish whether this is the case has 
>> not been done in 
>>> Australia.  Sound is a part of this - the majority of 
>> changes in owl taxonomy in 
>>> recent years have been based on calls.  How many recorded 
>> calls are there of lurida 
>>> boobooks to use as part of a taxonomic review?  As far as I 
>> know, just two, one by 
>>> Dave Stewart and one by Fred van Gessel.  If anyone else is 
>> aware of other 
>>> recordings, please let me know.  Of course for such 
>> recordings to be useful it 
>>> would be necessary to see (and ideally photograph) the 
>> calling bird as it is 
>>> necessary to confirm the recording is really of lurida.  
>> Also it should be noted 
>>> that calls don't seem to vary as much between species in 
>> the Ninox genus as they do 
>>> in some other owl groups.
>>> 
>>> Second, Cicadabirds.  The possibility of there being two 
>> species of Cicadabird in 
>>> Australia has been discussed for a while [e.g. Glenn 
>> Holmes, The Bird Observer, 
>>> 801; 12 (1999)].  While it has recently been suggested that 
>> this won't be resolved 
>>> properly without additional collecting of specimens [L. 
>> Jospeh, Emu 111(3) p. iii], 
>>> given that calls are a significant part of the puzzle, more 
>> progress could be made 
>>> if a lot of recordings were available to analyse.  If a 
>> sufficiently large sample 
>>> proved that certain call types were only heard in 
>> rainforest habitat and others 
>>> only in dryer woodland then it would tend to support there 
>> being two species.
>>> 
>>> I recommend www.earbirding.com as it has a lot of posts on 
>> calls of American birds 
>>> and shows the sort of information that can be obtained from 
>> sound recording.  For 
>>> example, the calls of a recent vagrant sparrow were 
>> compared to recordings from 
>>> various locations to see if the bird was from a migratory 
>> or resident population, 
>>> which was relevant to the chance of it being an escapee.  
>> There are also some 
>>> recommendations on equipment (see the equipment tab on the 
>> right side of the page).
>>> 
>>> Murray Lord
>>> Sydney
>>> 
>>> 
>>> 
>>> ===============================
>>> 
>>> To unsubscribe from this mailing list,
>>> send the message:
>>> unsubscribe
>>> (in the body of the message, with no Subject line)
>>> to: 
>>> 
>>> http://birding-aus.org
>>> ===============================
>> 
>> 
>> ===============================
>> 
>> To unsubscribe from this mailing list,
>> send the message:
>> unsubscribe
>> (in the body of the message, with no Subject line)
>> to: 
>> 
>> http://birding-aus.org
>> ===============================

===============================

To unsubscribe from this mailing list,
send the message:
unsubscribe
(in the body of the message, with no Subject line)
to: 

http://birding-aus.org
===============================

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU