On 05/01/2017 09:39, Jim Lesurf wrote:

snip

FWIW I've recently been transferring and de-clicking some Ellington 'Radio
Transcriptions' discs released on Decca London in the late 1970s.

Do you do it manually, in a sound file editor, and 'flatten the spikes'?
I've done that a few times, and the results are pretty good.

Or is there a decent software solution?


--
Cheers, Rob

In article ,
Iain Churches wrote:

"Dave Plowman (News)" wrote in message
...
In article ,
Jim Lesurf wrote:
Yes, I am familiar with this technique. It was known as dissolving.

Thanks. Previously, I'd taken for granted that in the era of tape
people would always use tape splicing. But maybe some people found
'dissolving' easier. When listening at first it sounded like odd
dropouts due to something like dirt on the tape. But when I looked at
the waveforms the thought came to me that it was a deliberate erasure.

In broadcast is was known as spot erasing. Some pro machines had this
facility - although more commonly used on one track of a multitrack.
With caution. ;-)


Spot erasing was a totally different thing, and used to
remove wrong notes or wrong beats (snare, BD, hi-hat, etc)
from one specific track on a multitrack machine. It left a
"hole" in the audio, which, in listening, was covered by
materal from other tracks. Spot erasure on a mono
or stereo tape, was, for obvious reasons, not an option.

Dissolving, a totally different technique, produced a cross fade
and was used exclusively on mono or stereo quarter in tapes, which
is what we are talking about here.

You mean effectively electronic editing? Please explain how this could be
use to remove clicks from an LP after transferring to tape?

--
*And the cardiologist' s diet: - If it tastes good spit it out.

Dave Plowman London SW
To e-mail, change noise into sound.

In article ,
Iain Churches wrote:
No self-respecting editor would want to do such work, so it was usually
given to trainees, who were instructed to "keep all the bits" (which
they did, numbered with white chinagraph pencil, and stuck to the front
of the tape machine with editing tape in the right order, until their
engineer or producer approved the job)

Ah. Forgot you never worked in the real world of broadcast. ;-)

No. Thought I have recorded countless project
that have been broadcast. But not quite the same
thing:-)

No it's not. You might have endless time available to do something. For
broadcast, the clock is usually ticking. If only for cost reasons.

When I was thinking about a career, I found that,
using three criteria, training, salary levels and
prospects, broadcast came right at the bottom
of the league table.

I'm glad you made the right choice for you, Iain. Others might enjoy the
challenge of working in broadcast. In much the same way as some may prefer
working on a live concert to recording a performance in a studio.

Besides, I wanted to work in a company
were things were done properly.
In the "real world of broadcast", your plexi
screens around drummers, and lapel mics stuck
to the bridges of violins with BluTack, were
clearly not optimum solutions:-)

You never attend live music events, then?

--
*A fool and his money can throw one hell of a party.

Dave Plowman London SW
To e-mail, change noise into sound.

In article ,
Iain Churches wrote:

"Dave Plowman (News)" wrote in message
...

The one and only time I've cut clicks out of an LP was on one even
the various broadcast libraries couldn't find another of. It was for an
ITV schools' broadcast with no budget to have it specially recorded.

Why did you not simply ask the record company for a 15ips Dolby A
tape copy from the master? These were always supplied very quickly
at no cost to broadcast. We used to send tapes to BH almost daily,
and even paid the courier:-)

Tee hee.

This would have saved the cost of your LP transcription, declicking,
leadering etc.

It's a pity you didn't do it properly.

Pity you don't have a clue about the LP in question. Or indeed the
timescale involved.

--
*Oh, what a tangled website we weave when first we practice *

Dave Plowman London SW
To e-mail, change noise into sound.

In article ,
Jim Lesurf wrote:
In article , Dave Plowman (News)
wrote:
The one and only time I've cut clicks out of an LP was on one even the
various broadcast libraries couldn't find another of. It was for an ITV
schools' broadcast with no budget to have it specially recorded. Won an
international Emmy too - but not for the music. ;-)

I make digital transcriptions for various reasons.

1) To remove clicks from old LPs.

2) To avoid the need to have to play the same LP again, risking added
wear to my ancient Shure styli, etc.

3) Convenience of being able to play the results in rooms where I don't
have the record deck.

Some of the second-hand LPs I bought are worn and so still sound lousy.
But others - after a careful declicking - sound very good. And I find it
easier to relax and enjoy the music when I'm not anticipating rifle shot
accompaniment.

Yes - all very valid. I did look up the date of the prog I was talking
about - 1986. So rather before digital audio workstations became common.
Although the company I worked for had bought its first AudioFile by then.
But that was fully occupied dubbing 'The Bill' ;-)

--
*Velcro - what a rip off!*

Dave Plowman London SW
To e-mail, change noise into sound.

In article , RJH
wrote:
On 05/01/2017 09:39, Jim Lesurf wrote:

snip

FWIW I've recently been transferring and de-clicking some Ellington
'Radio Transcriptions' discs released on Decca London in the late
1970s.

Do you do it manually, in a sound file editor, and 'flatten the spikes'?
I've done that a few times, and the results are pretty good.

Or is there a decent software solution?

Mostly I use the 'repair' function that Audacity provides. This is limited
to a max of 128 samples per channel. But in effect it examines the patterns
either side of the selected series and attempts to do a smooth
interpolation of the shapes.

Since I record the LPs as 96k/24 that means it is limited to problems that
aren't longer than about 1.3-ish ms.

That works fine for me in most cases. The main exceptions are much longer
crunches or bangs, or clicks that leave a long LF 'tail'. For them I might
accept doing a snip. Although I suspect they might be easier to fix if the
digital recording was done without RIAA I've never bothered. The above
works fine in most cases. for me.

In practice I find I rarely need to snip out a section or use some other
means. The 'repair' generally produces a result where I can't hear any
problem once I've chosen the right start and end points.

The main 'trick' I use is to use sox to generate a high-pass filtered
version of a recording. Usually second order with a turnover around 5kHz. I
then load that into Audacity alongside the recording to be declicked. The
filtered version helps some smaller clicks to stand out, so they act as a
guide to 'find the Lady' if in the full recording the click is hiding in
the audio waveforms. It also shows more clearly the HF departures from a
smooth shape, so aids deciding the start and end points for an optimal
repair.

Caution: If you do the above make sure to take care *not* to save the
result in a way that adds back in the hf filtered version! 8-]

There are automated ways to do this. However I'm happy with the above, and
it also serves as an excuse to have a close listen to the recordings as I
'work on them'. :-) So it is usually fun and an interesting challenge
rather than a chore. No-one is paying me, I'm just doing it because I
prefer to.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On 06/01/2017 15:17, Jim Lesurf wrote:
Mostly I use the 'repair' function that Audacity provides. This is limited
to a max of 128 samples per channel. But in effect it examines the patterns
either side of the selected series and attempts to do a smooth
interpolation of the shapes.

Since I record the LPs as 96k/24 that means it is limited to problems that
aren't longer than about 1.3-ish ms.

I use Adobe Audition and 88.2k/24 instead, but the principle is the
same. Audition has a sizeable licence fee, but I inherited a copy when a
project was canned.

88.2k not 96k as they are going to CDs, and I think the down-sampling
must be simpler.

Andy

On Sat, 7 Jan 2017 22:45:04 +0000, Vir Campestris
wrote:

On 06/01/2017 15:17, Jim Lesurf wrote:
Mostly I use the 'repair' function that Audacity provides. This is limited
to a max of 128 samples per channel. But in effect it examines the patterns
either side of the selected series and attempts to do a smooth
interpolation of the shapes.

Since I record the LPs as 96k/24 that means it is limited to problems that
aren't longer than about 1.3-ish ms.

I use Adobe Audition and 88.2k/24 instead, but the principle is the
same. Audition has a sizeable licence fee, but I inherited a copy when a
project was canned.

88.2k not 96k as they are going to CDs, and I think the down-sampling
must be simpler.

You'd think it might be simpler, but in fact the same algorithm is
used. But this is something Audition (inherited from its life as
CoolEdit Pro) does particularly well, with almost no artifacts.

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

In article , Don Pearce
wrote:
I use Adobe Audition and 88.2k/24 instead, but the principle is the
same. Audition has a sizeable licence fee, but I inherited a copy when
a project was canned.

88.2k not 96k as they are going to CDs, and I think the down-sampling
must be simpler.

You'd think it might be simpler, but in fact the same algorithm is used.
But this is something Audition (inherited from its life as CoolEdit Pro)
does particularly well, with almost no artifacts.

I wouldn't be surprised if both used the same resampling code as sox. This
is based on what used to be called the "Secret Rabbit Code" and can do
arbitrary rate conversions quite well. However I tend to use sox for rate
conversions, filtering, etc. Just use Audacity for dealing with clicks.

FWIW I did do some sample-by-sample comparisons on the results of using
Audacity to do this on 96k/24 files. And confirmed that the results were
identical except for the 'repaired' sections.

I'd probably also use 88.2k if the end-aim was Audio CD. But since I'm
leaving the results as 96k/24 (in flac) this isn't a concern for me.
However I'd agree that 2:1 ratio conversions are relatively simple to do
well. What may be unknown, though, is how a given ADC operates when
outputting different rates. Some may run at a high *fixed* rate and do
their own internal downsampling. In such cases you may be better off using
a sample rate for the capture that is a simple scale factor down from that
internal rate. Devil in the details.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On 01/06/2017 12:13 PM, Iain Churches wrote:
"Eiron" wrote in message
...
Six copies of Volume 5 currently available via eBay :-)

Two copies should be sufficient.
As the clicks will be in different places, you should be able to choose
the best bits of each. :-)


A cigar for that man:-)

Multiple copies of the original form the basis of
every good audio restoration project.

Question as a layman, I presume you have to phase lock or somehow
otherwise sync the two recordings very accurately. Or do people pick
the best recording and just splice the second track as and when required
to cover the worse bits?

Pete