Using Digital Tools to Repair Analog Audio Robert Siegel talks to Jamie Howarth about the next step in audio restoration: ridding analog-era sound of its inevitable speed variations by writing software that virtually recreates the original device on which a recording was made from the existing tape. The sound is then digitally fed back through that machine to correct errors.

Using Digital Tools to Repair Analog Audio

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


This is ALL THINGS CONSIDERED from NPR News. I'm Robert Siegel.

(Soundbite of ringing phone)

SIEGEL: Don't bother answering that. We're playing it for you because there's something wrong with that sound. And the story that we're about to present is about flaws like that in old recording technology, and how a clever new computer software program mends the errors of speed. That ringing telephone was a sound effect for a movie. It was recorded for a film use years ago. And over the years, it developed a case of acoustic jitters.

(Soundbite of ringing phone)

SIEGEL: Now, run that sound through a complicated virtual reconstruction of the old machine it was recorded on, and presto.

(Soundbite of ringing)

SIEGEL: Another example…

(Soundbite of train whistle)

SIEGEL: And the fix.

(Soundbite of train whistle)

SIEGEL: What some digital age smart guys have done is this: they've harnessed a silent signal that's on most magnetic recordings, and they've put that to work as a kind of clock. Then, they could tell their computer to tap its toe at that pace, and as the flawed, warbling tape or film was played back, it would obey that bias clock, as it's called, and presto…

(Soundbite of music)

SIEGEL: No tiny speed flaws. For example, this…

(Soundbite of music)

Unidentified Woman: (Singing) Shine away, you're glistening. Time starts with your choosing.

SIEGEL: Becomes this…

(Soundbite of song)

Unidentified Woman: (Singing) Shine away, you're glistening. Time starts with your choosing.

SIEGEL: My guest is Jamie Howarth. And Mr. Howarth, what is it you do, exactly?

Mr. JAMIE HOWARTH (President, Plangent Processes): What we've developed is a system that can remove speed variations that were caused by the motors and pulleys and spindles in film and tape.

SIEGEL: And how common are such problems in film and tape?

Mr. HOWARTH: To some degree, they are present in all analog tape, in all analog film.

SIEGEL: So that's a problem one would find with analog recordings that figured in most of my career in radio, which was if you recorded on tape, the tape can stretch whether it can get to the tape. What other sorts of problems are you finding here?

Mr. HOWARTH: Well, we also found that very, very fast instances of what's called wow and flutter manifest themselves as distortion. If the speed variation is quick enough, the perception doesn't present itself as a gargle or as a warble. If it's quick enough, it actually turns into a graininess, and a harshness that's omnipresent particularly in film recordings where there was a perforation along the edge of the film that created a very, very fast - 96 times a second - flutter that was heretofore impossible to remove.

SIEGEL: Now, you've sent us some examples of sound that had wow and flutter in it. What do you think is the best illustration of that?

Mr. HOWARTH: Well, I think the Vivaldi piece there, the NBC symphony orchestra piece is very good.

SIEGEL: Is this a disc or a reel-to-reel that we got in there?

Mr. HOWARTH: Now, this is a reel-to-reel. This is an air check from 1955.

(Soundbite of music)

SIEGEL: Boy, hearing that I can almost see the plastic reels on the tape recorder not quite turning through as they were recording on that day.

Mr. HOWARTH: We've offered a steak dinner to anybody who can listen to the entire piece.

(Soundbite of laughter)

SIEGEL: And the result after you make the playback machine operate at the true speed, you're telling me, of the flawed recording machine, you can make the NBC symphony sound like this.

(Soundbite of music from "NBC Symphony Orchestra")

SIEGEL: Well, you've done as much for the NBC symphony as Toscanini did right there.

(Soundbite of laughter)

Mr. HOWARTH: That's correct.

SIEGEL: It's not just a simple question of one speed being different. I mean, there are lags and all sorts of artifacts you're trying to…

Mr. HOWARTH: There are a variety of things overlaid on top of each other. You'll see a spindle that's operating at 400 bumps a second, and then something else is happening every half a second. It's a myriad, and it really is a miniature golf coarse of strange effects that the tape has to find its way through in order to be able make the recording. Those all superimpose.

SIEGEL: Now, you worked on it - I gather this is a part of your important work - on that the sound of the movie version of South Pacific, and I gather what you were working with was - it was a lot of wow here. And let's here it.

(Soundbite of music from movie, "South Pacific")

SIEGEL: That's a perfect example of the orchestra sounding very, very imperfect right there. What is the problem there? What's wrong with that recording of South Pacific?

Mr. HOWARTH: Well, the chemistry of film is such that the older films that were made with acetate break down into vinegar - acidic acid, if you will. And they stretch and warp. The warpage creates tremendous amount of strain, and the film will actually stop that occasionally and slow down or fight the machine.

SIEGEL: So once again, digitally, you have to go in and figure out what the orchestra meant at that moment?

Mr. HOWARTH: Essentially, yes. What we try to do is we try to track what the machine actually did.

SIEGEL: And let's hear the result.

(Soundbite of music from movie, "South Pacific")

Mr. HOWARTH: There's still a dropout where the film actually comes off the head, so you hear a little bit of a volume differential.

SIEGEL: To date, what is your proudest achievement in terms of restoring a recording?

Mr. HOWARTH: One of the proudest achievements is Leonard Bernstein conducting a great orchestra in L.A. in the early '50s on "On the Waterfront."

SIEGEL: Let's hear what it sounded like before it got the treatment.

(Soundbite of music from movie, "On the Waterfront")

Mr. HOWARTH: What we found in that was that the machine was doing a very, very slow kind of motion throughout. And it felt oceanic and unsteady. It's just queasy.

(Soundbite of music from movie, "On the Waterfront")

SIEGEL: The prescription is?

Mr. HOWARTH: Prescription is is to track it as closely as possible and to do a high fidelity restoration of the pitch control, which we, we hoped to have accomplished with this. I think we did fairly well.

SIEGEL: And let's hear that.

(Soundbite of music from movie, "On the Waterfront")

SIEGEL: But to use your imagery, it sounds a lot more anchored right now than it did before.

Mr. HOWARTH: Thank you. That's the point.

SIEGEL: You find this satisfying work, to bring these old analog recordings closer to what the people in the studio actually heard when they were first recording it?

Mr. HOWARTH: Oh, my goodness. Yes. I love this. I really do. It raises the hair on the back of my neck when we playback something that was previously unlistenable, and we can help that. That's really gratifying. And we're proud of it.

SIEGEL: Well, Jamie Howarth, thank you very much for talking with us, and for bringing us examples of what you do.

Mr. HOWARTH: Thank you, sir.

SIEGEL: Mr. Howarth is the president of Plangent Processes.

Copyright © 2007 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.