Topic: Best "hiss" reduction? - Help me save this recording!

[jj69]

I made a generally excellent recording of one of my favorite musicians.  Unfortunately, my gear let me down the night of the show.  The right channel of my Sony PCM-A10 failed (despite thoroughly testing it before leaving the house).  The channel was not completely dead, but what I wound up with is a recording with the right channel gain 22db lower than the left.  As I'm sure you can imagine, once the right channel was boosted 22db in post, so was the noise floor.  The PCM-A10 is a great sounding little recorder, but adding 22db of gain to a channel is going to be noticeable. 

The good news is this was a hard rock band, so the excessive noise in the right channel is not particularly noticeable for most of the show. However, during quiet passages, and particularly between songs, the left channel is dead quiet, while the right channel sounds like a bad 70s cassette recording without noise reduction. 

Can someone suggest the best available noise reduction workflow to repair this that will result in the least amount of annoying artifacts?  The software I have readily available is basically Adobe Audition 3.0 and iZotope RX7.  However, I'd be willing to invest in something better if it will provide a worthwhile result.  I've tried the Audition noise reduction in the past, and no matter what I did, I found that I always preferred the original recording with hiss over the annoying artifacts the processing would leave behind. 

Any suggestions greatly appreciated! 

[vanark]

Have you considered dumping the bad channel and duplicating the good one? there are little tricks to give it some sort of stereo feel. That kind of hiss can be tough to manage without interfering with the audio too.

[jj69]

Have you considered dumping the bad channel and duplicating the good one? there are little tricks to give it some sort of stereo feel. That kind of hiss can be tough to manage without interfering with the audio too.

No - I'd live with the hiss before going to fake stereo! 

[BlueSky71]

Dual mono can sound good pan left pan right etc. Whatever stereo image you may have at this point is probably gone anyways. What were you running, where and was it 007. if 007 get on with it and dump the right channel. Shit happens. I have had static on a channel before on a 007 situation, and was upset about it, until we went dual mono.

[robgronotte]

How do you know it is a problem with the recorder?  I have had this kind of problem before, and it was always the microphones.
Microphones can also act differently from night to night, but it's hard to imagine a digital recorder randomly not performing well.

[jj69]

How do you know it is a problem with the recorder?  I have had this kind of problem before, and it was always the microphones.
Microphones can also act differently from night to night, but it's hard to imagine a digital recorder randomly not performing well.

LOL.  Believe me, I am POSITIVE it's the PCM-A10.  I have been though seven (7) of them!  All of them have failed in exactly the same way - the right channel (through the input jack) dies, usually progressively over time (or in this case, without warning).  I've had two technicians disassemble and examine my failed A10's only to find nothing obviously amiss inside.  The theory is that whatever is failing is inside one of the chips on the board.  Me and one taper friend are the only people we can find who have experienced this problem (he has killed only 6 PCM-10's, so I'm ahead by 1!).  We use nearly identical gear.  The best theory we have is that the (uncommon) 12V battery boxes we use are somehow leaking voltage back into the PCM-A10's input jack and over time frying one of the chips.  The only reason we've persisted with this particular 12V box this long is because it is so tiny and convenient to use.  The same box has never been an issue with the Sony PCM-M10, Roland R-05 or R07 or any other PCM recorder.  It's probably a combination of a defect in the 12V box and an overly fragile chip in the A10. 

[nulldogmas]

Have you considered dumping the bad channel and duplicating the good one? there are little tricks to give it some sort of stereo feel. That kind of hiss can be tough to manage without interfering with the audio too.

If the hiss is in a specific frequency range, you could also try just copying that frequency range from the good channel to the bad one. It might sound weird, or it might sound great, hard to predict until you try it.

But RX has a great hiss remover (at least RX 9 does, not sure about 7), so try slamming that puppy into learn mode on a section that's just hiss, and see what it can do to the rest of that channel.

[robgronotte]

Yow, why do you keep using that recorder then? I have one and like it, but I imagine using a different one would work better for you.
I use a 12v battery with mine also, hope that doesn't become a problem for me. Who made your battery box?

[unidentified]

Sorry to hear about all the troubles you're having. It does make me glad that I stuck with my faithful PCM M10 Sonys rather than switching to the A-10. Why are you using a 12 volt battery box instead of a 9-volt one? I presume it must have to do with the mics you are running.

[jj69]

All good questions. 

We cling to the A10 for its small size, and particularly its rock-solid Bluetooth remote control connection.  The Roland R-07 Bluetooth remote is reliable only with iPhones - its Android performance has been practically useless from the day it was released.  Likewise, we cling to the 12V battery box for its tiny size and amazing form factor.  No other combination of recorder/batter box is even close to being as compact. 

I don't think 12V of power is necessarily required for the mics, but I have had zero luck with 9V battery boxes across the board.  First, they are MUCH larger than the box I prefer.  More importantly, every 9V box I have used has failed in the field every time I've tried it.  My experience is limited to the small Church 9V boxes - they all lose connection with the battery bumped or jarred in any way (and if the box is in your pocket, it will always get moved around). 

For the record, the 12V box I like is the Sound Professionals SPSB-4.  I was reluctant to identify it, because I don't want to disparage the manufacturer.  That box has never been an issue with any other recorder, and their other 12V boxes (like the SPSB-12) have never caused an issue with the A10 to date.  I also have not yet had the opportunity to contact SP and inform them of our issues.  My plan is to send them one of my SPSB-4 boxes if they're willing to test it. 

If you look at the photos of the SPSB-4, you'll see it's small size and its unique form factor and you'll understand why we like it.  It's compact and requires no wires for connection. 

For this recording, I'm going to try to try the recommended hiss removal in RX7.  If anyone knows if the hiss removal function is different (and significantly better) in newer versions of RX, please let me know if it's worth upgrading at this time! 

[goodcooker]

Sounds like you are dead set on using gear that doesn't work right or taking the best/easiest approach to fixing the borked recording you are working on presently so I'm not going to try to talk you out of either situation.

Make sure to run noise reduction on your file before increasing the gain. After increasing the gain do it again. I had to do this in several steps before - run noise redux, increase gain by 10dB, run noise redux again, add the rest of needed gain, run noise redux again.

If you have a stereo file that's only going to have an increased noise floor on one side your best bet is to split the file into two mono files and work on them separately - which is why I don't understand why you just wouldn't duplicate your known good channel and use some tricks to get stereo separation perception out of it. But, hey, it's your recording.

[unidentified]

All good questions. 

We cling to the A10 for its small size, and particularly its rock-solid Bluetooth remote control connection.  The Roland R-07 Bluetooth remote is reliable only with iPhones - its Android performance has been practically useless from the day it was released.  Likewise, we cling to the 12V battery box for its tiny size and amazing form factor.  No other combination of recorder/batter box is even close to being as compact. 

I don't think 12V of power is necessarily required for the mics, but I have had zero luck with 9V battery boxes across the board.  First, they are MUCH larger than the box I prefer.  More importantly, every 9V box I have used has failed in the field every time I've tried it.  My experience is limited to the small Church 9V boxes - they all lose connection with the battery bumped or jarred in any way (and if the box is in your pocket, it will always get moved around). 

For the record, the 12V box I like is the Sound Professionals SPSB-4.  I was reluctant to identify it, because I don't want to disparage the manufacturer.  That box has never been an issue with any other recorder, and their other 12V boxes (like the SPSB-12) have never caused an issue with the A10 to date.  I also have not yet had the opportunity to contact SP and inform them of our issues.  My plan is to send them one of my SPSB-4 boxes if they're willing to test it. 

If you look at the photos of the SPSB-4, you'll see it's small size and its unique form factor and you'll understand why we like it.  It's compact and requires no wires for connection. 

For this recording, I'm going to try to try the recommended hiss removal in RX7.  If anyone knows if the hiss removal function is different (and significantly better) in newer versions of RX, please let me know if it's worth upgrading at this time!

I do like the form factor of that battery box, although I have to wonder if having the admittedly small mass of that box attached directly to the input jack is adding stress to that jack, as opposed to the stress it would be added by flexible cord. But who knows. Best of luck in addressing the problem that you're encountering

[jefflester]

All good questions. 

We cling to the A10 for its small size, and particularly its rock-solid Bluetooth remote control connection.  The Roland R-07 Bluetooth remote is reliable only with iPhones - its Android performance has been practically useless from the day it was released.  Likewise, we cling to the 12V battery box for its tiny size and amazing form factor.  No other combination of recorder/batter box is even close to being as compact. 

I don't think 12V of power is necessarily required for the mics, but I have had zero luck with 9V battery boxes across the board.  First, they are MUCH larger than the box I prefer.  More importantly, every 9V box I have used has failed in the field every time I've tried it.  My experience is limited to the small Church 9V boxes - they all lose connection with the battery bumped or jarred in any way (and if the box is in your pocket, it will always get moved around). 

For the record, the 12V box I like is the Sound Professionals SPSB-4.  I was reluctant to identify it, because I don't want to disparage the manufacturer.  That box has never been an issue with any other recorder, and their other 12V boxes (like the SPSB-12) have never caused an issue with the A10 to date.  I also have not yet had the opportunity to contact SP and inform them of our issues.  My plan is to send them one of my SPSB-4 boxes if they're willing to test it. 

If you look at the photos of the SPSB-4, you'll see it's small size and its unique form factor and you'll understand why we like it.  It's compact and requires no wires for connection. 

For this recording, I'm going to try to try the recommended hiss removal in RX7.  If anyone knows if the hiss removal function is different (and significantly better) in newer versions of RX, please let me know if it's worth upgrading at this time!

I do like the form factor of that battery box, although I have to wonder if having the admittedly small mass of that box attached directly to the input jack is adding stress to that jack, as opposed to the stress it would be added by flexible cord. But who knows. Best of luck in addressing the problem that you're encountering

That seems like a much more likely culprit than a blocking capacitor failing.

[roffels]

Now you've got me paranoid about using my 12v battery box with my Sony A10... I don't  often but I don't want to go and ruin my gear.

[robgronotte]

I realized that I was confused earlier, I actually use a 9v battery.

[Rairun]

I'd try a pass of noise removal on the channel with hiss, and mix it with a duplicate of the good channel instead of boosting the volume all the way up. You will lose some stereo separation, but not all of it, and it might be enough to keep the noise down.

[moondust.and.solitude]

If you have a stereo file that's only going to have an increased noise floor on one side your best bet is to split the file into two mono files and work on them separately - which is why I don't understand why you just wouldn't duplicate your known good channel and use some tricks to get stereo separation perception out of it. But, hey, it's your recording.

I have to agree with this. I have lots of mono recordings from "way back" when buying a stereo recorder wasn't always an option. Lol. In transferring these oldies I have used this VST within Adobe Audition 3 and the results is impressive. It really gives it a stereo feel ...

"SHEPPi" Spatial Enhancer.

http://dallashodgson.info/articles/OpenAmbienceProject/

[fanofjam]

Not to pile on, but I've never had an issue with creating stereo in post either.  I don't really consider that fake stereo any more or less than having my two coincident mics flying in a venue is fake stereo...it's just two mics pointed in different directions.  For what we do, it's not like the right and left mics set up in coincident recording patterns are recording different sound sources.  The positioning of the mics causes the perception of right and left either due to differences in sound pressure or differences in sound timing...both of which can be created in post.  So whether stereo is created in post by futzing with two mono files to create differences in sound pressure and/or differences in timing, or by two mics in the venue, it can end up being really similar sounding. 

(Not to mention that some sound guys don't send stereo mixes to the PAs.)

[Gutbucket]

If the recording is important enough to you to put the post-production work into it, and if the noise isn't audible during most of the musical sections that are more energetic..

1) Duplicate the recording to additional channels in the DAW.  Keep the two copies identical for the time being and automate a cross fade between them just prior to each time the noise becomes objectionable, and back again to the original just after it becomes unnoticeable.  If you were to listen to the output at this point it would sound identical to the raw recording (the cross fades are being made back and forth between identical copies).

^You are now setup to try a few different things to fix the noise problem by applying whatever you do to only the copy that is cross-faded to during the noisy sections. Whatever treatment you apply will only affect the noisy parts of the recording, the parts where the noise is not perceptible will not be effected at all.

2) Try applying noise reduction in Rx.  Tweaking settings for minimal artifacts.  lf it works decently enough you are done.

If not, try instead..
3) Whatever other noise mitigation technique you want  (such as coping over only effected frequency region from the good channel).  If it works you are done.

(with either of the options above, if you can reduce the noise somewhat yet not sufficiently to keep it from being objectionable mostly because it is only on one side, you can try adding a little bit of noise to the clean side.  The noise will then at least sound balanced rather than lopsided.  Like a cassette.  Retro analog.  If acceptable, you are done.)

If not try instead..
4) Monoize to the quieter channel during the noisy sections.  The recording will cross-fade from noise-free mono during the quiet parts to stereo during the louder parts.  If acceptable you are done.

4a) Try applying pseudo-stereo techniques if you like.  The recording will cross-fade from pseudo-stereo during the quiet parts to stereo during the louder parts.  Tweak the pseudo-stereo with an ear toward minimizing the difference in spatial feel during the transition to and from real stereo, rather than seeking the most attractive pseudo-stereo effect on its own.