Re: [ecasound] extract, mix and save questions

From: Brad Fuller <brad@email-addr-hidden>
Date: Tue Dec 20 2005 - 04:18:00 EET

Jeremy Hall wrote:

>>>This is the second file, -y means offset into the file, -t says how long
>>>it should run.
>>>
>>>
>>thanks, jeremy.
>>At this point, foo2.raw is the extracted file that is 30seconds long,
>>right? I assume that you are saying that the file in step1 is 30sec long.
>>
>>BTW, Do you know if there is a way to have it start at the closest zero
>>crossing?
>>
>>
>>
>What is your definition of a zero crossing? If you would explain your
>task, I can help you figure out how to accomplish it.
>
>
When the waveform reaches zero -- the data is 0.
Cutting at an arbitrary point is bound to cut at a data point that is
non-zero and then when the extracted file is played back could create a
click.

>
>
>>>Next mix them down
>>>
>>>ecasound -a:1 -i source1.wav -y:1500 -a:2 -i foo2.wav -a:1,2 -o foo3.wav
>>>
>>>
>>I don't know why you used -y here. foo2.wav should be the length of the
>>first file (source1.wav). I guess the issue is how does ecasound mix the
>>two? does it bypass the WAV header and start at index 0 in the audio
>>data (I assume that it does). If that's the case, then wouldn't it be:
>>
>>ecasound -a:1 -i source1.wav -a:2 -i foo2.wav -a:1,2 -o foo3.wav
>>
>>Where source1.wav is the file in the first step: source1.wav and
>>foo2.wav should be the exact same length.
>>(also assuming there is a step before this that converts foo2.raw to
>>foo2.wav)
>>
>>
>>
>I used the -y flag after the first source to show you that you can combine
>some of the steps into one. You can even use ewf files and play segments
>like regions of a playlist (to use ardour terminology)
>
>The -t flag is a global chain operator and tells ecasound how long to run.
>The ewf filetype is powerful. If one aio-object (input) is shorter than
>the others, ecasound is supposed to feed a zeros loop so that you don't
>get uninitialized noise coming out of that chain.
>
>If you're trying to find a specific point in a track so you can precisely
>perform a cut, the interactive mode of ecasound is helpful because you can
>make it run up to the marker place you want and monitor with the -t flag
>to make sure you don't overshoot. At one time there was a buffering bug
>so that you'd get an off-by-one loop depending on your buffersize,
>annoying for sure, which is one reason I did the manual math method.
>
>
I need to run this in a script, so interactive mode would be out.
I'll have to read up on ewf files... I don't know what they are.

>If you're writing cd tracks, remember your filesize (minus wav header if
>you wish) must be divisible by 2352 bytes (4*44100/75)
>
>_J
>
>
>
>>>It is important to make sure you use the same srate and that you read
>>>through the requirements from going from mono to stereo etc--use the erc
>>>effect if you want mono in stereo mode.
>>>
>>>Is that what you want? NOTE ecasound does an implicit -ea:50 on each chain
>>>(because there are two, and 50+50 == 100) so you may need to adjust your
>>>coefficients during mixing.
>>>
>>>
>>>
>>>
>>That's good to know, and it's appropriate that it does that.
>>
>>
>>
>>>Hope this helps.
>>>
>>>_J
>>>
>>>In the new year, Brad Fuller wrote:
>>>
>>>
>>>
>>>
>>>>I couldn't find a way to do this in ecasound, but maybe there's a way to
>>>>do it. Here's what I want to do:
>>>>
>>>>Find the exact length of a audio file in ms. I think ecalength can do
>>>>this, I know it's not sample accurate, but I just need down to the
>>>>around 1ms. Maybe there's another option other than ecalength. IŽd like
>>>>to have it sample accurate, or at least to the nearest zero crossings.
>>>>
>>>>Then, I want to extract a section of another file: at a specific
>>>>location in ms and at a length of the file above. I see that sox may be
>>>>able to do this using the 'trim" feature.
>>>>
>>>>Then, mix that extracted segment with another file and save the result
>>>>to a new file.
>>>>
>>>>How can this be done? Any ideas?
>>>>
>>>>thanks,
>>>>
>>>>brad
>>>>
>>>>

-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Ecasound-list mailing list
Ecasound-list@email-addr-hidden
https://lists.sourceforge.net/lists/listinfo/ecasound-list
Received on Tue Dec 20 08:15:04 2005

This archive was generated by hypermail 2.1.8 : Tue Dec 20 2005 - 08:15:04 EET