Marks in constant recording

Alex Hayes pulls a phone out of his pocket Recently a few friends came up to the mountains for the TALO swap/meet 05. We talked about a variety of things to do with teaching and learning online. The nature of the conference was open source - meaning that there was no agenda, key note speakers, or otherwise structured communications. We just got together and started yaking! It was great. Some really good ideas and thoughts were flowing freely. We talked over a Mexican lunch, we talked while walking the National Pass Trail, we talked over drinks at the Carrington, we talked and talked.

Recording such conversation is almost impossible. Remembering all the great ideas is just as impossible. But I remember one idea Alex Hayes (pictured), Sean FitzGerald and I had that could solve this problem - the marking in constant record.

Take your average mp3 recorder. Mic yourself up and hit record. On a 512meg card you should expect a few hours of recording, enough to get the majority of a night conversation. But going back over that massive flog of raw recording to edit it down to digestible content is a job I wouldn't wish on anyone. On top of that, once the 512 card is full, you have to find another card to get the breakfast conversations. Its just too much. What we really need is a recorder that does constantly record, but when you have that moment in conversation where you realise, "hey, that's a great idea!" you reach down to the mic cable and press a button that marks near to the end of the sample you want to keep. The recorder stores your out point mark and adds an inpoint mark automatically - say 5 minutes before the out. When the cards nears full, the recorder deletes everything but the marked recordings, freeing up some card to continue recording. Back in the editing misery, at least you know that the majority of what you have is what you though was good at the time. You might miss a few things with that auto 5 minute in point mark, but I'd say on the whole you'd get most of it. 5 minutes should be enough I'd imagine, for someone recording to realise what was being said was worth keeping. If that person was to mark an out point within 5 minutes of the last, then the reorder would know that more than 5 minutes was needed, it would just keep the in point at the original mark point.

This could work for video as well of course.

Creative Commons License

This work is licensed under a Creative Commons Attribution 2.5 License.


Blogger gary.l said...

Hi Leigh

This sounds exactly like what I need. My thinking is that in digital photography pictures are put into folders. With your idea the idea of pressing a button woth the preceesing 5 minutes cut and paste into specified folders. Anything data/recording thats not put into the folder can then be written over once all the un used spaced has been spent.

With the advent of large ipod type devices holding hard drives of 40GB and more, the amount of hours of recording could be endless.

Blogger alexanderhayes said...

My response was;

[www. alexanderhayes.com - October]

I often find myself mid-sentence wondering why, with all the available technologies around that I continue to choose to lose the best ideas to time.

TALO2005 gave me some time to contemplate what must be done to podcast the best of what I speak of and the interaction I have with others without the formal accesories of setting up a podcasting mainframe, enrolling a zillions microphones and spending countless hours editing and tidying up on what is essentially ( and mainly with me) blue sky theories.

Much like tagging, the AudioMarkMaker should employ a mechanism that records live - at least a nights worth of content. That means if I choose to leave the recording device live the whole time I'm out from adjusting my tie to the moment I pull off my dacs before I hop into the shower I could continue to record.

5 hours of stereo with a good bit rate should be possible.

Leigh Blackall , Maria Tevaskis, Micheal Nelson , Sean FitzGerald , Tony Lorriman, Steven Parker , Anne Paterson, Don Perrin , Stephan Ridgway, Alex Hayes , Peter Shanks , Helen McFadden, Jenny Hartlett all contributed to the idea.

The device type should be small, affordable and discrete. The mircrophone should be directional, lapel aesthetic, non-intrusive to the wearer and sensitive to all that transpires within a 3 metre radius. The audio recorder;

1. Records continuously till prompted to 'mark'

2. Upon prompt [ button depress or better still audio prompt] the device runs back 5 - 10 mins and 'marks' in a start point and runs back to the original 'mark'. The action of capturing the mark should not interefere with the continuous recording nor should it be heard, seen nor apparent to the wearer.

3. The device provides all necessary elements for immediate export including duration, file size, times to pre-set calendar and so on. Nothing should be left to 'step ins' - ie. device should boot instantly.

4. When the device is presented to export recordings in either bluetooth, USB or preferably wi-fi format it masks all recorded content outside of marked enclosures and files these intermediates into one recording, effectively compiling untimed edits as one recording.......handy if something pertinent in the preceeding leadup was essential and needs reclaiming.

5. The device should be rechargable - direct power - provide shuffle power for listening immediately and have the ability for onboard audio levels adjusting. Ipod type style ear jacks should be enabled.

The benefits of this device to me are immense. When attending 'AudioMarkMaker' approved events I would have sustainable dialogue with fellow delegates without the traditional conventions of shoving podcasting enable devices into someones face. To date the mobile phone has been my preferred 'mobcasting' device where i can author audio to the net for immediate RSS aggregation.

I would have the ability to discretely edit the tripe and shite out of my recordings with minimal time wastage.

Those impromptu 'strikeups' with unknowns suddenly become legitamite research fodder. Having discussed this at length with some of the TALO group makes me think that such a device would be incredibly important to any number or individuals whose interests in social dialogue are often pensive till the warmup period is over and then they let fly in 5 - 10 minute bursts like me, accelerated often by interjections, arguments, points of clarification and other social nicities.

Lets face it. I'm no talker.

In fact when in esteemed company I like to listen more than talk and this might be a handy tool when I am proximate to brilliance yet loath to make a fool of myself. The automated abilities of the device might prove handy also when I'm climbing into a conversation from a differing angle and posthaste gives me leverage to approach divergent thinkers whose ideas and brilliant interludes become fuzzy over recall through a nice red wine.

Rebooting my computer and editing for another twenty four hours bores me and hence my need for you to solve my immediate desire. Build me it and tell what I need to pay for it. Maybe we'll swap our intellectual copyright for your efforts as we engage another generation of learners with the new architecture you have so valiantly fought to design.

Smiles. Show us whats possible.

Alex Hayes
October 2006

Blogger nswtox said...

Sounds like an excellent idea.

Blogger Mark said...

I've seen a device somewhere that has a tiny video camera attached to a pair of sunglasses that constantly records 30 seconds, and then records over it to record the the next 30 seconds.
You have a button that you press to capture/save the last 30 sec recording onto a flash drive so you can to review later.

I'm not sure if it was ever made or not...sounds perfect for Youtube though.


Post a Comment

<< Home