Narrating Blues

[KateESal]

Paul, I hugely sympathise...losing work that you've spent several hours on is the WORST.

You can use Reaper for free forever if you want (the evaluation version is fully functional, you just have to wait a few seconds for a reminder notice to disappear once you've finished the evaluation period).


But to purchase Reaper, you only need to make a one-off payment, you don't need to take out a subscription. And it's incredibly well-priced, especially given the level of functionality you get. Including a raft of built-in virtual equipment, like compressors, de-essers and the like, to help sort out fixable sound issues and enhance your recordings.

I paid for Reaper about eight years ago and it's been through numerous upgrades and things since, but I've never been asked for more money and have never lost any existing work.

There are a few differences in the editing process compared with Audacity, but it won't take you long to get your head round them...plenty of tutorials online, too.

Most importantly, you can set up automatic back-ups and suchlike so you hopefully won't lose big chunks of work, should there be any glitches.

There are some elements of fine editing that Audacity is better for than Reaper. But you can set up Audacity to link with Reaper so you can use its editing capability, should you ever need to.

I hope Audacity will be able to restore your work. Once you've got it back, I recommend you export it all in .wav format and switch to Reaper.

Good luck Paul, I really hope they can recover the recordings.

 

[Paul Whybrow]

Thanks for the advice, Rich. I should have been more cautious, but in learning how to use Audacity, mainly through YouTube videos, I saw no mention of the need to save work and use Audacity as just a workstation. Nevertheless, I saved a copy on a memory stick, but in the inaccessible aup. file. format I could try opening it, but might get the missing data file message again, whose remedy then destroys the project!

The thing is, it doesn't always do this, as before losing Chapters 1 & 2 I converted the Opening and Closing Credits and the Retail Audio Sample into MP3s with no problem. They're all small files, though, which might be a factor.

A way around the problem would be to make multiple copies of each chapter (known as a project in Audacity), before trying to convert a chapter into an MP3 to upload to ACX. How do I do that without opening it? I use Linux, so I'm going to have to find out. Any ideas @AgentPete or any other Linux users?

What annoys me about this 'bug' is that worldwide, there are thousands of writers and musicians who are going to be losing work. Should I publicise this error?! The developers sure as hell is hot won't be mentioning it!

I'm doubtful that the developers I'm in touch with will find a quick solution, and if they do come up with something to prevent data loss when a project is saved, will it be applicable to my 64 recorded and saved projects?

Looking online, there are lots of advice articles and videos about recovering lost projects, but these are mainly for when Audacity crashes (which never happened to me) rather than for Audacity doing the dirty on users.

I'm going to have to re-record and re-master the first two chapters, which is work I didn't need. I was intimidated by the challenge of turning Book 2, The Perfect Murderer into an audiobook, as it's so damned long at 140,000 words, but I forced my way through by working 10-16 hour days. This made me feel wretched, but I accomplished a little more each day. I trusted the software to work. I was wrong to do that.

Who or what can you trust these days?

 

[Paul Whybrow]

As a separate issue applicable to Audacity and maybe to Reaper and other DAWs, I found a remedy for some inexplicable noise interference on recordings.

As I mentioned above, for as long as I've been using Audacity I've been plagued by a noise I named Krik. I came around to thinking that it was a bug that appeared in Version 2.2.1-1, as it's absent (so far) from 2.4.2. Then, new noise pollution appeared in 2.4.2, a buzzing that reminded me of static feedback from the days of recording on tape. The waveform on Audacity was tiny and jagged.

My recording setup is that the USB microphone is powered by a cable connected to my laptop sitting two feet away. The mic has a jack port for connecting my earbuds to, so I can hear what I'm saying while narrating. When listening to playback I'd left this arrangement in place. Sometimes I'd get popping noises too, but they only sometimes happened, as listening to the recording again they'd be gone. After watching a YouTube video where the expert removed static from his microphone by moving the laptop further away, I experimented with my setup.

Distancing the laptop didn't make a difference, but I found that listening to playback with my earbuds connected to the laptop eliminated the static buzzing! It wasn't on the recording at all, but somehow created by the earbuds plugged into the mic.

Jeez, recording an audiobook is complicated!

 

[KateESal]

Hi Paul

Not sure if you're already doing this, but I would always export your work as a .wav file (or equivalent) rather than leave it sitting in the project format...even before you complete the project. Then you can open it with numerous other applications if things go wrong with the original app you're using.

(I should probably mention that Reaper will save everything you record in a .wav form prior to editing (as long as don't tell it NOT to save!) so you don't end up with a format that's specific to Reaper only. The project format is specific to Reaper (as it is to every DAW), but the component sound files are all accessible by other apps.)

As for buzzes and the like, electro-magnetic fields can often produce interference on recordings...fluorescent lights are a particular problem. And yes, it might only be the headphones that are causing the issue (it's a good idea to listen to your recordings on loudspeaker, if you can).

Yes, sound engineering certainly is a complex business!! What we can achieve in a domestic setting is amazing nowadays, but we also have to figure out ways of compensating for the lack of bespoke studio environment when we're doing it!

 

[Rich.]

I second all of that.

Knowing your gear, and knowing its limitations is important (if time consuming!). Audacity is a free open source program developed and maintained by a group of volunteers. With that in mind it's hard to see it negatively.

But the most important thing in this whole discussion is the importance of backing up your work as uncompressed WAVs, whatever DAW you use. The importance of backups cannot be overstressed!

 

[Paul Whybrow]

For anyone thinking of trying Reaper on Linux, this video will help you install it:

 

[Paul Whybrow]

This video made by a new user of Reaper reassured me that it's the way to go:

 

[Paul Whybrow]

Hmm, the plot thickens. I've been exchanging emails with Audacity developer James Crook, sending him data files for the two deleted chapters to see if he can rebuild them. It turns out that he can't, as there are too many block files missing.

The appearance of my 2.4.2 shocked him, as it wasn't right, and he asked me from where I'd downloaded it. I didn't know that software distributors built their own version of apps....thinking that they were supplied directly from the developers. I got 2.4.2 from Filepuma, which seemed to be OK, though I've since found that reviews are variable:

Is filepuma.com Safe? Community Reviews | WoT (Web of Trust)

What I'm still uncertain of, is if I install a safe version of Audacity will it be able to open the stored projects without erasing them? Or, have they all been infected with the bug, from my editing them?

I've downloaded Reaper and have been confusing myself watching YouTube videos of how to use it.

I'm being cheeky, but @KateESal what settings do you use? There are more buttons on Reaper's console than a Jumbo Jet!

 

[Paul Whybrow]

Six months into narrating, I’ve learnt a lot—much of it, stuff I didn’t want to know! I’ll have said some of this in previous posts, but I hope that it helps anyone starting out on creating their own audiobook.

* Everything takes much, much longer to do than you think it will. At least six times longer and probably ten times longer.

Why? Because so much is out of your control. However carefully you’ve chosen your recording equipment, there will still be strange pops and hisses on your recording. As for noise pollution from further away, you may not be aware of it until you listen to playback zzzzz when zzzzz you’ll zzzzz hear a zzzzz moped zzzzz whine past. Grasping at straws, you’ll erase the racket from spaces between words, but it’s still discernible underneath words. You’ll think for a moment, “I can away with that”, before deciding that you really can’t, so it means re-recording and re-mastering

* If you wrote it, you’ve got to narrate it. Recording my second audiobook, The Perfect Murderer, which was the first Cornish Detective novel I wrote, I’ve literally been cursing myself aloud (while mastering) for going on so much. That book was originally 180,000 words long, which I edited down to 140,000 words. It’s still too long! One chapter lasts 44 minutes!

If you intend to create an audiobook, keep chapters to no longer than twenty minutes. It will be so much easier on you and your larynx and your patience and the reader’s patience.

It’s the same with awkward words. Such as, if using a foreign name make it one that’s easy to pronounce! The Perfect Murderer contains Polish and Serbian names, which has caused me to twist my tongue. zly uroc for evil eye is OK, but what about zmiennoksztaltny' for shape-shifter?

* As I’ve recently chronicled, software can let you down, so take the time to back up what you’ve recorded in safe places. Because of using a version of Audacity which had been incorrectly built by the shareware site Filepuma, I lost 250 hours of work.

I placed too much faith in Audacity’s fine reputation, unaware that shoddy versions of it existed.


*Do yourself a favour by not having too many characters talking in a scene. You’re going to have to change your voice to indicate who is talking which is tricky and has the potential to sound like a parody if you lay on the accent too thickly.

A way around this is to record each character’s lines separately, then splice them together—easily done digitally.


* Don’t agonise if you make a small mistake, such as saying ‘pretty’ instead of ‘beautiful’. Narrating an audiobook isn’t a test of accuracy. Some listeners following the story in the book may notice, but so what? Say it was altered when editing.


* Use the punch and roll technique when you make a mishtake. Immediately repeat the phrase that you mucked up, rather than thinking you’ll come back to it later to correct it. You won’t sound the same then. You’ll love yourself for having done so, as it makes editing so much easier.

Using Punch and Record for audiobooks


* Get used to the fact that you won’t feel like a writer when narrating and mastering your recordings. For me, it’s not an enjoyable process, more like being an audioslave.

 

[Paul Whybrow]

I've just opened each of the 64 chapters to find that 21 are ruined. Opening them produced a warning-error message about missing block files and orphan files, offering me three choices of how to carry on—the only way to close the project is to choose one of them—each of the options erased the project. Imagine what that's like...hundreds of hours of work destroyed because a distributor of Audacity built a shoddy version of it.

I'm going to email the developers again to find out if they've done anything to make Filepuma take down their 2.4.2 distro. Looking, I see that it's still on their site. How many more writers and musicians are going get fucked over?

I'm going to have to re-narrate and re-master the 21 chapters. Goodbye the next two weeks.

Too much grindstone at the moment.

 

[AgentPete]

Thanks for the advice, Rich. I should have been more cautious, but in learning how to use Audacity, mainly through YouTube videos, I saw no mention of the need to save work and use Audacity as just a workstation. Nevertheless, I saved a copy on a memory stick, but in the inaccessible aup. file. format I could try opening it, but might get the missing data file message again, whose remedy then destroys the project!

The thing is, it doesn't always do this, as before losing Chapters 1 & 2 I converted the Opening and Closing Credits and the Retail Audio Sample into MP3s with no problem. They're all small files, though, which might be a factor.

A way around the problem would be to make multiple copies of each chapter (known as a project in Audacity), before trying to convert a chapter into an MP3 to upload to ACX. How do I do that without opening it? I use Linux, so I'm going to have to find out. Any ideas @AgentPete or any other Linux users?

As everyone else is saying here, Paul, I totally sympathise. Horrible experience.

I only have very casual experience of Audacity, so am out of my depth here. At least one of the coders is in touch with you… can/will they assist in file rescue?

When I was doing a lot of audio work, I went from Pro Tools to Reaper, as @KateESal mentioned, which is now my standard go-to audio editor. But also, one of the two reasons I still have a Windows machine… Reaper on Linux is “experimental”, so be careful.

 

[Paul Whybrow]

Thank you for your support, one and all. The gods continue to use me as their plaything. As stubborn as mule (as intelligent as one, too!), I decided to make an immediate start on re-recording the destroyed chapters using a brand new official version of Audacity. I may get around to learning Reaper, but I'm going with what I know at the moment.

I opened Audacity 2.4.2 and set the recording level, to avoid distortion. This had the effect of freezing the app and the laptop. Nothing responded, so I powered off. Rebooting, I tried again...with the same result, a frozen laptop. Imagine my joy! Rebooting, I went to serve myself dinner, a frugal meal of baked potatoes and sardines in tomato sauce. Adding extra tomato sauce from a squeezy bottle, the flip cap was jammed in place. I gave it a tug, which expelled a large dollop of cold sauce onto my stomach and groin!

I heard the gods laugh!

 

[Emily]

Paul, I think you really could make a fortune as a YouTube Vlogger. I know this is every kind of horror for you, but really, thinking positively, it's comedy gold.

 

[Hannah F]

Paul, I think you really could make a fortune as a YouTube Vlogger. I know this is every kind of horror for you, but really, thinking positively, it's comedy gold.

I agree. This would also make a great sitcom. (Though that maybe not what you want to hear.)

 

[Paul Whybrow]

@Rich. you said something intriguing above: You can also use a similar punch in technique during post production, which will allow you to surgically replace sentences or even individual words.

How do you do this?

 

[Paul Whybrow]

I'm beginning to think that I'm going insane!

Last night, I made a start at rebuilding my second audiobook by re-recording five of the destroyed chapters. I saved them in wav and MP3 in several places, including Dropbox.

Returning to the task this morning, I recorded another chapter, only to be confronted with a new warning message, this time about Dropouts. The warning looked inaccurate, as I use an SSD with 979 MB of space left and my memory stick has 27 GB. I put it down to a glitch, and saved my project before seeing if it would reopen.

Once again, there was a warning-error message about missing audio blocks!

I've emailed developer James Crook (is his surname apt?!) explaining the situation in a strongly-worded message.

At present, I am creating work for Audacity to ruin.

Watch this space...which contains me in a padded cell!

 

[Rich.]

@Rich. you said something intriguing above: You can also use a similar punch in technique during post production, which will allow you to surgically replace sentences or even individual words.

How do you do this?

It's a little fiddly. You create a second track, copy and paste to it some seconds of audio from before the cut you want to make, and then perform a punch and roll on that bit of audio on the second track. You then copy and paste the newly recorded audio back into track one in the appropriate place, and delete the thing you're replacing. You can do amazing things with this – right down to editing syllables. Before you perform the copy and paste, make sure you press z for At Zero Crossings. This will reduce enormously clicky cuts.

Let me know if this (rather rushed) explanation makes sense

 

[Paul Whybrow]

Update: Corresponding with developer James Crook today has led to a position where neither of us know what's causing Audacity to destroy my projects!

His conclusion is: "Oceanaudio is free https://download.cnet.com/Ocenaudio/3000-2170_4-75567510.html and may be worth trying. Please don't use Audacity for your production work anymore. It's obviously not working for you on the system you have."

Implying that it's my system that's at fault makes no sense, as I used Audacity to record 80,000 words of the first book & 140,000 words of the second book without altering any settings on my system. Then, it went feral on me and started eating itself!

I'm now in the position of having a ruined audiobook, with two-thirds of the chapters OK (at the moment...will they stay whole?) and the other 21 chapters needing to be re-recorded in Reaper or Oceanaudio.

Somehow, I've become the only person in the world for whom Audacity doesn't work.

If I give up on the entire malarkey, I'll feel worse than if I continue, but only just.

 

[Paul Whybrow]

As I use Linux Mint, I had a look for DAWs designed to work with Linux.

Ocenaudio was well-reviewed, so I'm going to try it out. It looks simple enough for my frazzled brain to handle:

Oceanaudio An Audio Editor For Linux

 

[KateESal]

Oh god Paul, what a nightmare you're going through...! Yes, an editing system designed to work specifically with Linux is a good idea. As Pete pointed out above, Reaper's compatibility with Linux doesn't seem to be fully known at this stage, so it would probably be as well to pause that idea for the time being (the last thing you want is another glitch messing up your hard work).

In answer to your question upthread, the only aspect of Reaper which I fiddle with ahead of recording are the volume input levels: each individual track has a volume control and there's a master volume control too. That said, I use a Behringer U-phoria sound interface and I usually twiddle the physical controls on that to get the levels about right. I then arm the track to record by clicking on its individual red button and when I'm ready to roll, hit the big red master record control which starts the recording process. When I'm finished, I press the space bar and when prompted, I click yes to save the recording. I then get on with editing.

If I make a mistake during the recording, I simply repeat the offending phrase from a suitable point during the recording and edit out the cock-up afterwards. If I only notice a mistake when listening back, I do what @Rich. mentioned and open a second track, arm that to record (having dis-armed the previous track, so I don't accidentally record on top my first track...although it's not a disaster if I do, Reaper has a good undo feature) and record the correction in parallel with the original. Then I cut out the chunk of botched track from the first recording and drag the correction up to fill its place (you can drag the edges of the recordings together so it all fits nicely). It sounds complicated, but actually it's a relatively straightforward process, once you've done it a couple of times.

I then add a couple of FX to the track: a de-esser and a compressor usually. Both are plugins that come with the software, you simply choose them from the FX list and apply a preset (or create your own settings, if you prefer).

And then I export the final result as a .wav. (in Reaper, it's called "rendering" the recording). I convert the .wav to an .mp3 using Apple Music (what used to be iTunes), but there are various freely available apps which can do this conversion.

Good luck with Oceanaudio, I hope it's the answer to your recording problems!

(And Emily's right, you've found a whole new area for blogging...)

 

[Victoria Bastedo]

Amazing, Paul, thank you. From reading this one wonders, how to paid narrators make any profit? They must have to put in hours more of time than they get paid for. I considered narrating my own books, but I'm intimidated and find the process confusing. Thank you for putting up the info videos!

 

[Rich.]

From reading this one wonders, how to paid narrators make any profit?

Most of them don't, I have discovered (the hard way).

 

[AgentPete]

As I use Linux Mint, I had a look for DAWs designed to work with Linux.

Paul, just a word on this (no doubt unneeded). DAW refers to the machine, not the program / app. My first lesson in audio was to have a computer entirely dedicated to one thing – audio. This machine does not connect to the internet. You do not update the operating system. The whole thing is essentially frozen once you’ve set it up and it’s demonstrably stable. That machine is then your DAW, and you can rely on it.

Fortunately, and unlike video, you don’t need any great processing power for this. A modest second-hand machine will usually do just fine.

Also, you may already have checked out Ubuntu Studio, which is already pretty much set up (even then, I’d remove all the unnecessary apps).

 

[Paul Whybrow]

Paul, just a word on this (no doubt unneeded). DAW refers to the machine, not the program / app. My first lesson in audio was to have a computer entirely dedicated to one thing – audio. This machine does not connect to the internet. You do not update the operating system. The whole thing is essentially frozen once you’ve set it up and it’s demonstrably stable. That machine is then your DAW, and you can rely on it.

Fortunately, and unlike video, you don’t need any great processing power for this. A modest second-hand machine will usually do just fine.

Also, you may already have checked out Ubuntu Studio, which is already pretty much set up (even then, I’d remove all the unnecessary apps).

@AgentPete, do you think it would make recording an audiobook easier and the result more professional if I buy a Digital Audio Workstation? Some of them look complicated, while others have few controls, such as this one:

Amazon product ASIN B07QR6Z1JB
Amazon Prime Days are nigh, so I might get a deal that will make my life easier as a narrator and amateur sound engineer. I have no desire to become expert in this area, it's more looking for ways to eliminate the dreaded phantom noises that appear in a recording. I know I'm not causing them, but in Ocenaudio as with Audacity, I hear 'KRIK' underlying my narration. Theoretically, it's possible to remove them by adjusting decibel filters, but that's time-consuming and it mutes rather than destroys them. I may be too much of a perfectionist, but I've wasted hundreds of hours re-recording sections with noise pollution.

I was wondering if a DAW would better block the strange noises.

I'll be trying Ubuntu Studio. Thank you for recommending it.

 

[Rich.]

Paul, the Scarlett is an interface rarther than a DAW. It allows you to connect professional microphones (the ones with XLR connectors rather than USB) to your PC. I'm not sure that using them makes things any easier, but they do allow for more professional quality. They don't "block" strange noises, but they may generate fewer artefacts (unwanted noises created by the equipment).

 

[Paul Whybrow]

Artefacts, @Rich. ? Reminds me of my days as a librarian and cataloguing and storing artefacts. I remain bemused by the phantom sound interference I get, which has me on the edge of my seat waiting for it to appear. It's become my Horla.

Hours spent looking for solutions in forums and on YouTube have only yielded the usual advice about removing background chaff, the sort of hissing and whirring that happens on any recording.

What's really strange, is that KRIK often makes an appearance in the first two minutes of recording, then goes into hiding. I need an audio exorcist!

 

[Rich.]

Yeah, tracing that kind of problem is a screaming headache, and usually involves going through everything piece by piece (which I'm sure you've done!). After that, you put it down to gremlins (and smash it with a hammer).

A point of clarification...

Paul, just a word on this (no doubt unneeded). DAW refers to the machine, not the program / app. My first lesson in audio was to have a computer entirely dedicated to one thing – audio. This machine does not connect to the internet. You do not update the operating system. The whole thing is essentially frozen once you’ve set it up and it’s demonstrably stable. That machine is then your DAW, and you can rely on it.

...while this is absolutely true in a professional environment, the reality for many non-professionals is that your DAW also doubles up as your general home PC (it certainly does in my case). You make the best of it – or if all else fails, you take out the aforementioned hammer.

 

[AgentPete]

The quality of the audio interface itself is usually not the limiting factor (with a few exceptions, see below).

Many of them have pre-amps built in, such as the Scarlett Solo you mentioned. Again, as long as the pre-amp is reasonably colourless, which most of them are at that price range, you can add any additional processing (e.g. equalisation) during post-production. Aim for a neutral recording, and then add whatever colour / quality you want during post-production. Reaper has a great range of free plugins such as EQ, compression etc, which may be all you need. I use a few iZotope plugins, not free but very good.

The limiting factor is almost always a combo of the mike itself, your own mike technique, and the room dynamics. The latter is a very interesting rabbit hole to explore

As far as the Scarlett Solo is concerned, I have a strong aversion to it! I tried it once for Pop-Ups, and it blew up. I had to do that show on a mike headset, which sounded horrible. I know, it can happen to any device, but it really put me off that one.

The Apollo Twin gave me even more problems, altho I do like it’s amazing sound quality… when it works. It’s essentially a Mac animal, since it uses a Thunderbolt interface. Getting a Thunderbolt interface to work properly on PC audio is non-trivial. Never did for me, had to be sent back. And as for Linux, forget it.

I used to have a ton of outboard gear, now reluctantly sold because in truth, you can do almost all of that stuff in the box now.

Currently using an Audient iD14 pre-amp and interface. No problems so far. Not sure re Linux status, you might have to ask on forums to see who’s using it and how it’s doing.

 

[AgentPete]

@Paul Whybrow re artefacts... post a link to an example somewhere (e.g. YouTube). Let us hear what you're getting. We may be able to help...

 

[KateESal]

I'd add to Peter's warning to avoid the Scarlett...I haven't had much luck with those interfaces wither. I have a very reasonably priced Behringer U-phoria audio interface and I've been much happier with it.

It's worth having a reasonable audio interface (the Behringer's only about fifty quid or so) and a decent-quality mic (again, there are some well-priced models that will do a perfectly good job).