digitalFAQ.com Forum

digitalFAQ.com Forum (https://www.digitalfaq.com/forum/)
-   Restore, Filter, Improve Quality (https://www.digitalfaq.com/forum/video-restore/)
-   -   Is this tape salvageable? (https://www.digitalfaq.com/forum/video-restore/12763-tape-salvageable.html)

Major17Wood 05-09-2022 04:18 PM

Is this tape salvageable?
 
1 Attachment(s)
Horrible buzz throughout the tape, it does almost completely go away sometimes though. Any way to fix it?

Attachment 15169

dpalomaki 05-09-2022 05:30 PM

1 Attachment(s)
You could use noise reduction filter in a program such as Adobe Audition and it would make a significant improvement. It will take a bit of trial and error to get the optimum noise sample and filter setting. And it may be necessary to process the audio in pieces.

FWIW: The noise is a 60 Hz spatter with harmonics at 60Hz intervals up to around 1 kHz (see attached waveforms sample.) . The variation in amplitude may be caused by camcorder audio AGC and/or orientation of the camcorder and its leads to the noise source.

timtape 05-09-2022 08:01 PM

I'm unable to access the file. Any chance you could upload a basic wav audio file? Or an mp4 file?

From the look of it, it is a mistracking HiFi sound track.

The proper way is to attempt the best playback from the tape, avoiding the artefacts in the first place. At the least, minimising them. That can then bring the problem down to proportions that audio software can handle.

Probably the better software tool is not a Denoiser but a Declicker, perhaps also a Decrackler. I demonstrated this recently on a poster's example. Here's the thread:

http://www.digitalfaq.com/forum/vide...-fix-hifi.html

But this minimising can be beyond most people as it can involve playing the tape on a range of VCR's until the best result is obtained, including custom adjusting some VCR parameters such as manual tracking and tape height guides.

If it is a HiFi soundtrack, there is usually another completely separate audio track on the tape called Linear or Normal audio. It's lower quality sound but when the HiFi track is bad and cant be repaired enough, the lower quality Linear track can be far preferable. Some consumer decks will automatically default to the HiFi track in which case a deck capable of playing the Linear track only may be needed.

Cheers Tim.

dpalomaki 05-09-2022 08:54 PM

1 Attachment(s)
The attached file is a sample the result of application of restoration>noise reduction twice and then 10 dB gain using Audition. This was a quick and dirty with a single random noise sample and no attempt to refine it. Not sure what the background sound was supposed to be. This may be close, and it might be way off. (I suppose it could pass for too loud music spun by a DJ at a dance/reception.)

It is interesting to se in the waveform the spatter spikes are of opposite polarity in the L and R channels, but the audio between spikes is pretty much in phase as if mono, just at a different recorded level.

The video appears to track OK and given the fixed relationship of video to HiFi heads the audio should be tracking OK as well. I wonder if the strobes are causing it?

How and when was the original recorded?

The linear track audio may tell the story.

Good luck with the project.

timtape 05-09-2022 09:41 PM

Hi dpalomaki, thanks for uploading the wav file. Would you mind also uploading the unprocessed audio file, please?

Most video editing programs probably contain the generic audio Denoiser but not the more targetted tools like DeClick and Decrackle . I was hoping to try these tools on the OP's file if possible.

With audio restoration tools, there is a specific "order of processing". Denoising comes at the end of the process, not the beginning, if at all.

As per this teaching article:

https://www.izotope.com/en/learn/ord...perations.html

Tim.

Major17Wood 05-09-2022 10:30 PM

Quote:

Originally Posted by dpalomaki (Post 84689)
The attached file is a sample the result of application of restoration>noise reduction twice and then 10 dB gain using Audition. This was a quick and dirty with a single random noise sample and no attempt to refine it. Not sure what the background sound was supposed to be. This may be close, and it might be way off. (I suppose it could pass for too loud music spun by a DJ at a dance/reception.)

It is interesting to se in the waveform the spatter spikes are of opposite polarity in the L and R channels, but the audio between spikes is pretty much in phase as if mono, just at a different recorded level.

The video appears to track OK and given the fixed relationship of video to HiFi heads the audio should be tracking OK as well. I wonder if the strobes are causing it?

How and when was the original recorded?

The linear track audio may tell the story.

Good luck with the project.



It was made by a production company and sold to kids who went to the prom. probably crudely dubbed as they needed to make hundreds of copies.

dpalomaki 05-10-2022 06:44 AM

2 Attachment(s)
Have you tried playback in another VCR?

In what year was the event?

Given the wave form, where in the spikes of noise approached clipping and the wanted audio was perhaps 10-to-20 dB down taming the spikes was a first priority. Denoise is often thought of as removing background things like hum, hiss and HVAC noise. In this case the spatter is the dominant audio component.

It sort of reminds me of an issue caused vertical blanking spatter in audio of broadcast (or Channel 3/4 output from a VCR) when the video brightness substantially exceeds 100 IRE. To some extent the audio waveform also looks a bit like it is timed to the variations in the top center strobe light.

Raw audio attached and the full wave form of the raw audio clip.

timtape 05-10-2022 08:31 AM

3 Attachment(s)
Thanks dpalomaki for the unprocessed file.

Here are the relevent files, renamed for clarity.

Even though a big improvement this is a difficult case. A better playback off the tape would probably yield a better final result.

dpalomaki 05-10-2022 09:21 AM

Declick did good. While not a great original recording, the bottom line for the OP is that the audio probably can be substantially salvaged to become listenable at least for those invested in the event.

Major17Wood 05-10-2022 02:26 PM

Quote:

Originally Posted by timtape (Post 84699)
Thanks dpalomaki for the unprocessed file.

Here are the relevent files, renamed for clarity.

Even though a big improvement this is a difficult case. A better playback off the tape would probably yield a better final result.

That result is much better than telling future viewers of the video to turn the sound off :)
How difficult is it to run declick on my entire video?

Also, Im not gonna get any better playback than this. Im using a HS-HD2000U which plays my other tapes perfectly, its just this one tape that sounds like trash for some reason.

Major17Wood 05-10-2022 02:32 PM

Quote:

Originally Posted by dpalomaki (Post 84701)
Declick did good. While not a great original recording, the bottom line for the OP is that the audio probably can be substantially salvaged to become listenable at least for those invested in the event.

Yeah I'm just trying to make it listenable. I could post as is, but I'd kinda be embarrassed to.

timtape 05-10-2022 05:42 PM

2 Attachment(s)
Quote:

Originally Posted by Major17Wood (Post 84711)
That result is much better than telling future viewers of the video to turn the sound off :)
How difficult is it to run declick on my entire video?...

The entire video is longer so it just takes the computer longer to process. The essential first step is choosing the correct audio tool for the job and then adjusting it for best results.

Maybe the simplest way is done in the video editing program using the Declicker tool as an effect on the audio track. Alternatively the audio track can be exported to an audio editing program, Declicked then imported back into the video program.

As dpalomaki noticed, most of the music is in the centre of the speakers (mono) and the remaining noise on the left and right edges, so summing to mono reduces the noise a little more. With the noise now a lot lower, the music can be turned up without nasty distortion. Sample attached with original for comparison.

Major17Wood 05-10-2022 06:40 PM

Which audio program are you using and which version of de-click do you have? Could you also post your settings?
I've installed Audacity with this version of de-click https://forum.audacityteam.org/viewt...245549#p245549
but I'm having trouble getting an acceptable result.

timtape 05-10-2022 07:53 PM

Quote:

Originally Posted by Major17Wood (Post 84719)
Which audio program are you using and which version of de-click do you have? Could you also post your settings?
I've installed Audacity with this version of de-click https://forum.audacityteam.org/viewt...245549#p245549
but I'm having trouble getting an acceptable result.

For your sample I used the Declicker tool in Izotope RX 8 Elements. $50 US at the moment I think, included with various other audio production tools in "The Soundwide Intro Bundle". * Not sure the exact settings I used but unlike with Audacity the settings can be adjusted while listening which I find a lot easier than in Audacity.

* Note these are VST plugins only. They work only as an effect in a compatible existing video or audio program, hence the lower price.

Izotope also provides free time limited demo software. After a time it still works but limited functionality.

There are other products from various companies.

The digital Declicker tool is not new. It was first used professionally in the late 1980's. It made possible quality reissues on CD of historic 78 RPM recordings. Most people associate the Declicker with 78 and vinyl records. But as you can hear it can do good things with other formats too like clicky HiFi video sound tracks.


All times are GMT -5. The time now is 04:25 PM

Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.