Re: [linux-audio-dev] peakfiles and EDL's

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: Re: [linux-audio-dev] peakfiles and EDL's
From: Robert Schrem (Robert.Schrem_AT_WiredMinds.de)
Date: Mon Feb 26 2001 - 11:13:44 EET


On Sat, 24 Feb 2001, Paul Davis wrote:
> Design Question Time.
>
> consider the following situation. you are using pre-generated
> peakfiles, containing the max/min amplitudes for every N samples, to
> dislay waveforms. and you're also using an EDL for editing. lets make
> N=2048 for easy thinking.

Before takeing every 2048 sample for the peak file I think the
audio data should be filtered for at least two reasons:

1. Avoid alising introduced by very low frequencies.
   Alised samples wouldn't be representative anyway.

2. The Peak data should represent the loudness of the corespoding
   region in the HiRes file.

If the peakfile only holds 'loudness' values we can compute linear
combinations of peak file data in any disired combination. If there
is a time offset between two peak file maps we just pick the nearest
neigbour of each peak file. We don't have to think about phase aspects.
If the user zooms into the time domain of the recording, we will have
to switch from peak file data to the raw data to be able to show all
details.

> a user has edited the audio data stream at a point that is 3074
> samples into one file, by inserting another file at that point. you
> now have to recompute the waveform display. You get the first value
> for the first 2048 samples. The next peak value needs to be based on
> the 1024 samples from the first file and the succeeding 1024 samples
> from the next file.
>
> But wait: the peak values are precomputed, and may represent max/min
> values that are not in the sample data that now forms the audio data
> stream (e.g. the min value for the first file's 2nd block of 2048
> might occur at sample 3786). So how can we possibly decide what
> max/min values to use for the 2nd chunk of 2048 samples in the audio
> stream ? Its presumably based on both files, but can we determine it
> without reading the actual audio data for that part of the audio
> stream ?
>
> And it gets worse: what happens if the inserted material is not
> aligned with the first sample of the second file, but is offset. Now,
> every precomputed max/min pair for this file is essentially useless
> because they are "out of phase" with the way the audio is actually
> being used.
>
> Solutions or even suggestions welcome.
>
> --p


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Mon Feb 26 2001 - 11:46:53 EET