Production Expert

View Original

Todd-AO Release Dialog Restoration App For Just $49

Todd-AO have released Absentia DX with an algorithm that is designed to analyze production dialog recordings and then remove obvious hums, wireless rings, and ticks, while maintaining the integrity of the human voice. Scroll down for before and after examples.

Todd-AO was founded in 1953 as a joint venture of Michael Todd and the American Optical Company for the purpose of developing and distributing a large format presentation system incorporating a wide, curved screen with multi-channel sound. Over the years Todd-AO has evolved into a post production sound company and now they have developed this app.

We understand that ABDX was originally developed for a network television show with difficult production sound that resulted in substantial repetitive manual labour and now they have made it available for all of us to use. 

Todd-AO tell us that you can simply drag and drop volumes, folders, or sound files directly onto the application or the settings window and files will begin processing. An Absentia DX progress window will appear, with the number of sound files queued, and a status bar of the files being simultaneously processed.

ABDX sound files’ metadata and Sound File ID are identical to the original files except for the noises that have been removed. With identical metadata you can swap the original files with ABDX files or vice versa. This will mitigate the fact it is a stand alone app and will save a lot of time having to replace media or persuading Pro Tools to point at the processed files.

Perhaps the best bit is the price, just $49. Initially it is only available as a Mac OSX app but a Windows version is apparently coming soon.

There is also a 3 day trial available, so why not download it and try it on some files and let us know what you think of it.

Quick Test With Before & After Examples

We have downloaded the trial version and I have tried it on some of my stock test files and so here are some before and after clips.

Your browser doesn't support HTML5 audio

No Earth

This first example was a recording made on a minidisc recorder at a funeral. There wasn't time to check the equipment beforehand and so when it came to play it back there was alot of hum and buzz because the screen had become detached in the mic cable.

Your browser doesn't support HTML5 audio

No Earth Processed

The fix is pretty spectacular, all the harmonics have gone. I had the Hum Remover option enabled with the Strong option as well as the Broadband Reducer option enabled.

Your browser doesn't support HTML5 audio

Tape Motor

This was a recording made on an old open reel tape machine by two childen recording a little drama they had written. It used the internal mic and is plagued with tape motor noise.

Your browser doesn't support HTML5 audio

Tape Motor Processed

Again the processed version is very good. I used the broadband Reducer and Hum Remover options and it has handled the fact that the motor noise isn't a constant frequency.

Your browser doesn't support HTML5 audio

Hiss

His is a recording on a minidisc recorder that when the mic lead isn't wired correctly produces a high level of broadband hiss, and is quite a challenge to clean up.

Your browser doesn't support HTML5 audio

Hiss Processed

This has made a small improvement to the his level but isn't anywhere as spectacular as the first two examples.

Your browser doesn't support HTML5 audio

Scratched Vinyl Record

This is a scratched vinyl record recording that I use to test the de-click function on audio restoration software.

Your browser doesn't support HTML5 audio

Scratched Vinyl Record Processed V1

I ran this using the Broadband Reducer, Hum Remover and Tick Remover options and as you can hear, it has done something strange to the file. So I ran it again and below you can hear the processed file with just the Tick Remover option enabled.

Your browser doesn't support HTML5 audio

Scratched Vinyl Record Processed V2

As you can hear this has fixed the clicks on the recording very well.

Conclusion

So in conclusion from these quick tests it would appear to work well on the Tick Remover and Hum Remover options but less well on the Broadband Reducer, but this maybe that the algorithm is set for background ambience rather than hiss.

We will look into a more detailed review and test the metadata features as well.

See this gallery in the original post