Techwatch: Recovering sound from silent film/video

[Rodney]

This reminds me of a few far fetched story ideas I wanted to use in a project many years ago.

In those projects I intentionally targeted things that were 'impossible' because the technology added an element of mystery, fantasy, science and fiction and suggested a sense of the futuristic as a point of departure within an otherwise present day context. The trick being to tie the impossible tech into reality just enough to keep it from being entirely unbelievable with the gist of the resolution of the plot being something of a Rube Goldberg approach to problem solving; "How did the hero solve the mystery?" Why, Everyone knows he's a master of forms of science and technology very few people have mastered of course!

 

It would seem that those ideas are as far fetched as this tech (and moreso) but the crazy thing is that I'm sure most folks would consider sound recovery from silent imagery just as impossible.

 

 

And this doesn't even begin to approach what can (theoretically) be done when sound is also captured at the time of recorded imagery and then compared and contrasted with the imagery. And what of lip reading... and other tech that can be brought to bear to recreate what no one knew was there in the first place.

 

In related tech, it wasn't that long ago that someone sought to recreate the sounds from the early 1900s captured in a room where clay was being formed on a lathing wheel.

Yet another related piece of tech being that of recording light itself in slow motion as it bounces off of surfaces to reveal things not in direct view (shades of Bladerunner).

 

Are you safe whispering to a confidant in a room full of people jabbering with loud music playing? Well, that might depend on what else is captured in the video. Whatever you do don't stand too close to that plant... and avoid eating potato chips out of a bag as you do!

[robcat2075]

So every potato chip bag is now an eavesdropping device.

 

It's hard to believe they can recover that from a video image but the sound is so awful... well, maybe. If that really works then no above-ground room is safe for confidential discussions.

 

I once read about a spy technique that involved bouncing a laser off a window pane to read the vibrations caused by the sound in the room. Sort of the same notion.

 

I remember a story about Henry Kissinger. He was in Moscow negotiating an arms treaty and at some point the Russians asked for a copy of a document he was showing them. Just to make the point that he knew they had hidden cameras in the conference room he held the paper up to a chandelier and shouted, "Three copies, please!"

[Rodney]

There is a reason recording devices of any type are not allowed in sensitive locations.

But stuff like this suggests that even Maxwell Smart and Agent 99 wouldn't be able to securely communicate inside their cone of silence.

The joke there of course being that they couldn't even hear themselves while in the cone of silence.

 

[NancyGormezano]

Fascinating! Nice find.

[robcat2075]

Those old-style laugh tracks ruin old comedies.

[markw]

Very interesting find Rodney.
I couldn't help but wonder during the video how loud was the sound coming from the speaker? Is there a threshold below which this will not work, there then being insufficient energy in the sound wave?

[robcat2075]

I couldn't help but wonder during the video how loud was the sound coming from the speaker?

 

That s a good point. If they could have made this work with just a person talking nearby they would have done it.

[Simon Edmondson]

A while back a friend and I were in conversation about using lasers to record sound. The initial idea was to use them to record drums or percussive instruments. The light would project up inside the drum and be reflected back off a prepared surface, picked up and processed. It would have the advantage that, because the only moving part was the drum surface, it would not "blow out' the ribbon used in a standard Mic. There would be all sorts of other advantages in terms of data processing possibilities as well. We went on to develop the idea further so that it might be used on guitars to pick up the light reflected off the strings and, by varying the light position on the body of the instrument, you could develop a much broader tonal spectrum than the standard positions of 'normal' guitars.

 

We looked at trying it out, and he had someone who was 'interested' but we never took it further than that really. Other than looking at possible kit to use.

The devil is in the detail... !

simon