bioacoustics-l
[Top] [All Lists]

Re: Record multiple crickets simultaneously

To:
Subject: Re: Record multiple crickets simultaneously
From: Gianni Pavan <>
Date: Wed, 15 Nov 2006 23:38:33 +0100
Dear Susan
I think the best option is to set a PC based recording system. It is flexible, programmable, and a lot cheaper than stand-alone recorders. We developed a software to record simultaneously and continuously up to 48 channels in a single multichannel file. We could easily adapt it to record individual files when the signal is above a threshold. Otherwise you could run several instances of our sound spectrograph; in such a case each copy of the software records a timestamped stereo file everytime in one of the two channels there is signal above a threshold in a specific frequency band. In case all 50 crickets will sing together I'm not sure a single pc will be able to record to disk 25 stereo files at the same time; in such a case it could be safer to split the task on 2 or more separate low cost PCs.

We worked well with multiple MOTU firewire interfaces: for example the HD896 costs less than 1500$ (verify in US) and provides 8 channels with very good mic preamplifiers with 96kHz sampling. If you need to spend less, the PRESONUS FIREPOD offer similar features for 600$ (verify in US). By using PCI multichannel boards such as the MAUDIO 1010 you could further halve the cost but then you need to add external microphone preamplifiers.

Otherwise you could use a software made by an italian company with our cooperation that allows to setup a network (100mbit or 1Gbit) with multiple PCs each one equipped with sound boards. Each PC is a "sender" connected to a "director" PC that receives through the network the data streams coming from all the sound boards installed in the "senders"; the director allows to connect each channel from "senders" to "recorders" , logical devices on the "director" PC that can be programmed to record a file each time the signal of a channel goes beyond a threshold in a specified frequency range (in the future version it will be possible to setup an autoadaptative threshold that takes into account slow variations of the background noise). Recorded files have a timestamp in the filename and the name of the sender/channel. You may use, say, 8 PCs each one with an 8 channel sound board to feed 64 individual recorders on the "director" PC.
But you can also use less PCs with more sound boards each.
Probably, with 32 or 48 kHz sampling, you can also concentrate everything on a single PC with multiple disks on which to distribute the channels to be recorded (recording 64 channels together in a single file is easy, recording 64 independent files on the same disk is more difficult).

Mics could be cheap but good condenser microphones with P48 powering that may cost about 30$ each (I think at Behringer C2). You may spend a lot less if you build your own condenser mics with capsules such as the Primo model 158.

Gianni Pavan

--------------------------------------------------------------
Gianni Pavan
Centro Interdisciplinare di Bioacustica e Ricerche Ambientali
Universita' degli Studi di Pavia
Via Taramelli 24, 27100 PAVIA, ITALIA
Tel         +39-0382-987874
Fax        +39-02-700-32921
Web       http://www.unipv.it/cibra
Email      


<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 Bioacoustics-L 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