Fabric Rub on Lav Mic Fix ?

[Ken Beals]

Didn't intentionally bury the mic. A few posts earlier on this thread I share basically what happened.

In brief when arrived to gig I was told there would be another Speaker in front of the key Speaker.

Had I known I would have brought a 2nd wireless to properly set the key Speaker so it would be seamless transition.

When transferred the wireless to key I put transmitter pack into inside breast pocket attaching mic to lapel as it should be.

The Speaker pulled it out put in his front pants pocket. I had a piece of gaff tape holding the excess mic cable still stuck from previous Speaker and now a severely shortened cable.

Since was using the Rycote stickie I stuck it on his shirt. When walked away he buttoned up his jacket burying the mic.

So it was a chain reaction of fatal mis-steps and the pressure to rush it . Ultimately it was still my fault and now simply looking for answers/help and knowing in future will not let this happen again.

[Greg Miller]

Ken,

I've been playing with one little area (0:27.35 - 0:29.0 "a shift within them") for about 45 minutes now. I can not get a result that I like, using any form of filtering or noise reduction. The noise starts before the "ft" in "shift" ends, and I cannot successfully remove the noise without compromising the end of "shift." Also, after I filter out all the noise, then there is a very obviously filtered "hole" in the audio, between "shift" and "within."

If I look at a spectral view of this section, I cannot visually discern where the noise starts. If I run a brute-force lowpass filter, I hear noise until I get the filter frequency down around 750 Hz. The noise is very broad band, and therefore not something that can be filtered out easily or unobtrusively.

If I try the adaptive noise filter, the result is worse than I can achieve manually. If I try an impulse noise filter (hoping that the "scrape" is a series of "clicks") the result is worse than I can achieve manually.

The best solution I can find is to replace the word "shift" with another "shift" from a different part of the file, plus some filtering on the syllable "with." Even so, the result is obviously artificial to my ear. It's a painstaking and exhausting process. And you might not get this lucky with some of the other noise hits. You won't find replacement words for every hit, and the duration of some of the hits spans multiple words.

I've asked twice how long the clip is, and you have never replied. So let's assume it's 30 minutes. I counted eleven noise hits in the minute-long sample you posted. Multiply by 30, that's over 300 noise hits. This seems like an incredibly huge project to tackle. If it were the last known recording of, for example, JFK, MLK Jr., FDR, or Winston Churchill, it might be worth the effort. But...

[Ken Beals]

Greg - thank you for trying - apologize for not responding about duration est. 20 min.
Hard lesson will have to apologize to the one who hired me and forfeit getting paid - an expensive un-pro bumble on my part - thanks all who posted

[Greg Miller]

Ken,

Someone else might have better luck, or find a way to do it faster. Unfortunately, I tend to look for "broadcast quality" results which makes everything very demanding. It's hard to make myself quit, and move on to the next repair, when I can still hear problems. Even so, if we assume there are perhaps 200 noise hits, that's a painful (and perhaps impossible) thing to tackle. I hate to be the only one chiming in here; I would love to hear someone else's assessment of the situation.

I don't know the nature of this shoot, or why your client wants the video. If it's just for archival purposes, then the monologue is intelligible despite the noise. If they want to release a DVD for sale, that requires a different level of quality.

I agree that it ultimately is your responsibility but IMHO it's the presenter's fault for moving the equipment after you had placed it, and also partially the client's fault for failing to tell you in advance about the second presenter. (Then again, it wouldn't hurt to carry a spare.) As someone else mentioned, it's a shame there wasn't a backup recording from the PA system; I assume you've inquired about that. Hindsight is always easy.

If I were in your shoes (given that I don't really know the details) I'd probably explain clearly that the presenter caused the problem, then give the client the raw footage gratis, as a "peace offering." I hope you can resolve it amicably.

[Jim Andrada]

I gave myself 20 minute to play with the first rub in Izotope - not perfect but see what you think

https://www.dropbox.com/s/ixf24s41yx...irst%20try.wav

[Tom Morrow]

That doesn't sound to me like fabric rub either, more like a RF hit. Although if the mic had fallen off the fabric completely and was banging around somewhere that could be it.

I don't like Rycote stickies; they aren't sticky enough in most situations. Marginal stickiness can be noisier than a bad placement as the poorly sticking material makes velcro noises as it sticks and resticks while the clothing moves.

Good question about whether to interrupt the person. That really comes down to a judgement call based on what you know about how important the recording will be. The hard situation is when the client overstates the importance of the recording and then is annoyed when you obsess to get it right.