temporal cut/paste editing using amplitude

Nov 29, 2006 at 6:30pm

temporal cut/paste editing using amplitude

Hey everyone. Sorry about the long mail, but I’ve been thinking about
this patch for a while, and I’m still a bit stumped about where to
start. Thanks in advance for any help…

OK, basic gist:

I want a patch that takes a full res video with sound (varying
lengths, usually an hour or two), then gives it back to me with all
the parts that have audio above or below a certain threshold removed
(and the gaps in between closed). The output must have synched sound
from the original.

So I guess jit.record in realtime is out, with the sound deal.

I kind of have an idea that I might use spigot~ (which I can’t seem
to get to work on my MacBook Pro, Max/MSP v 4.6.2 and Jitter v. 1.6.2
- but will delay that question until synthesisters.com and
cycling74.com are back online, so I can search) and jit.vcr with
peakamp, but I wanted to know:

1. is there a way to have jit.vcr add on to the end, rather than
saving separate files?
2. if not, any better ideas?
3. if not, I figured I’d use sprintf to write files like movie1.mov,
movie2.mov, etc – up till how ever many (probably several hundred or
thousand). Anybody know how to easily stitch these together, either
in Jitter or another application.
4. Probably an easy one: what would the message be to jit.vcr (or
jit.record, or whatever you recommend) for the right sound/video
compression that would not require rendering in Final Cut Pro (NTSC
and PAL)?

Bonus Question:
1. Any ideas on how I might adjust said patch to give a “minimum
length” of each splice? This is not absolutely necessary, but it was
something to think about… peakamp’s speed and sample rate both come
to mind, but I worry I might miss something rather than give a
minimum length (which still might be ok, I’ll only see when I test
this, but I wanted to get a few thoughts from the experts before I
get started).

Thanks so much again in advance!

Best,

nathaniel

http://nathanielstern.com

#28967
Nov 29, 2006 at 8:59pm

this is what i’d do: i’d play through the movie in realtime,
checking audio with spigot~. when the amplitude cracks the level,
i’d stop the movie and use the editing features of jit.qt.movie to
cut the splice out and make a new movie, and save it with audio and
video. no recompression needed, thus no worries about video format.
and it’s pretty easy to munge your splice sizes this way- just set
your editing method to frames or qt units or ms.

use the same editing features to splice together all of your
resulting small movies.

so, which movie is it? you’re putting together all of the noisy
parts of action movies now? cool!

On Nov 29, 2006, at 1:30 PM, nathaniel wrote:

> Hey everyone. Sorry about the long mail, but I’ve been thinking
> about this patch for a while, and I’m still a bit stumped about
> where to start. Thanks in advance for any help…
>
> OK, basic gist:
>
> I want a patch that takes a full res video with sound (varying
> lengths, usually an hour or two), then gives it back to me with all
> the parts that have audio above or below a certain threshold
> removed (and the gaps in between closed). The output must have
> synched sound from the original.
>
> So I guess jit.record in realtime is out, with the sound deal.
>
> I kind of have an idea that I might use spigot~ (which I can’t seem
> to get to work on my MacBook Pro, Max/MSP v 4.6.2 and Jitter v.
> 1.6.2 – but will delay that question until synthesisters.com and
> cycling74.com are back online, so I can search) and jit.vcr with
> peakamp, but I wanted to know:
>
> 1. is there a way to have jit.vcr add on to the end, rather than
> saving separate files?
> 2. if not, any better ideas?
> 3. if not, I figured I’d use sprintf to write files like
> movie1.mov, movie2.mov, etc – up till how ever many (probably
> several hundred or thousand). Anybody know how to easily stitch
> these together, either in Jitter or another application.
> 4. Probably an easy one: what would the message be to jit.vcr (or
> jit.record, or whatever you recommend) for the right sound/video
> compression that would not require rendering in Final Cut Pro (NTSC
> and PAL)?
>
> Bonus Question:
> 1. Any ideas on how I might adjust said patch to give a “minimum
> length” of each splice? This is not absolutely necessary, but it
> was something to think about… peakamp’s speed and sample rate
> both come to mind, but I worry I might miss something rather than
> give a minimum length (which still might be ok, I’ll only see when
> I test this, but I wanted to get a few thoughts from the experts
> before I get started).
>
> Thanks so much again in advance!
>
> Best,
>
> nathaniel
> http://nathanielstern.com
>
>
>

#89552
Nov 29, 2006 at 11:15pm

Wow. I’ve never even noticed the “editing” subpatcher in jit.qt.movie
before – I am really gonna have to play a bit now. Awesome. Thanks Josh.

In theory, I understand what you are saying, but I think I need to
play with this a bit. I’m wondering if rather than exporting splices,
I could just go through the movie, getting start and end points of
the splices I DON’T want, saving the movie as I go along with the
automation.

The tricky part will probably be getting both the start AND end
points of any given splice, then cutting it. I guess this could be
done with spigot~, peakamp~, togedge, and a whole lot of messages,
right? This could be easier than I thought…. Moderate a min time
with a splice by just setting a min time based on the difference
between my start and end… Oh, I’m liking this. THANKS!

Sound feasible?

Oh yeh, and good call Josh. Your show has started my brain on a whole
lot of diff “language lapse” ideas (most of which have to be done by
hand – this idea being one exception). Am starting with an
installation of 6 “quiet narratives” – the silent (or at least very
very low volume) parts of loud movies (or movies with mostly a
soundtrack playing)… Will being work this week, as well as on
another hand-done edit…

Thanks!

nathaniel

http://nathanielstern.com

On Nov 29, 2006, at 8:59 PM, joshua goldberg wrote:

> this is what i’d do: i’d play through the movie in realtime,
> checking audio with spigot~. when the amplitude cracks the level,
> i’d stop the movie and use the editing features of jit.qt.movie to
> cut the splice out and make a new movie, and save it with audio and
> video. no recompression needed, thus no worries about video
> format. and it’s pretty easy to munge your splice sizes this way-
> just set your editing method to frames or qt units or ms.
>
> use the same editing features to splice together all of your
> resulting small movies.
>
> so, which movie is it? you’re putting together all of the noisy
> parts of action movies now? cool!
>
> On Nov 29, 2006, at 1:30 PM, nathaniel wrote:
>
>> Hey everyone. Sorry about the long mail, but I’ve been thinking
>> about this patch for a while, and I’m still a bit stumped about
>> where to start. Thanks in advance for any help…
>>
>> OK, basic gist:
>>
>> I want a patch that takes a full res video with sound (varying
>> lengths, usually an hour or two), then gives it back to me with
>> all the parts that have audio above or below a certain threshold
>> removed (and the gaps in between closed). The output must have
>> synched sound from the original.
>>
>> So I guess jit.record in realtime is out, with the sound deal.
>>
>> I kind of have an idea that I might use spigot~ (which I can’t
>> seem to get to work on my MacBook Pro, Max/MSP v 4.6.2 and Jitter
>> v. 1.6.2 – but will delay that question until synthesisters.com
>> and cycling74.com are back online, so I can search) and jit.vcr
>> with peakamp, but I wanted to know:
>>
>> 1. is there a way to have jit.vcr add on to the end, rather than
>> saving separate files?
>> 2. if not, any better ideas?
>> 3. if not, I figured I’d use sprintf to write files like
>> movie1.mov, movie2.mov, etc – up till how ever many (probably
>> several hundred or thousand). Anybody know how to easily stitch
>> these together, either in Jitter or another application.
>> 4. Probably an easy one: what would the message be to jit.vcr (or
>> jit.record, or whatever you recommend) for the right sound/video
>> compression that would not require rendering in Final Cut Pro
>> (NTSC and PAL)?
>>
>> Bonus Question:
>> 1. Any ideas on how I might adjust said patch to give a “minimum
>> length” of each splice? This is not absolutely necessary, but it
>> was something to think about… peakamp’s speed and sample rate
>> both come to mind, but I worry I might miss something rather than
>> give a minimum length (which still might be ok, I’ll only see when
>> I test this, but I wanted to get a few thoughts from the experts
>> before I get started).
>>
>> Thanks so much again in advance!
>>
>> Best,
>>
>> nathaniel
>> http://nathanielstern.com
>>
>>
>>
>

#89553

You must be logged in to reply to this topic.