Quality Assurance Meets Absurd Quantity

We all need differentiators to help set us apart from the growing army of voiceoverists* so that we end up with our fair share of the sacks of money* waiting to be spent on voiceovers.

(*Inside joke from the VO-BB.com)

One of the things I ”sell” is my quality assurance. I spend time making sure that the files I send are as close to perfection as possible. This is fairly easy to do if you have a standard :30 or :60 spot. Couple of minutes to record a couple of takes. A quick couple of minutes to listen and clean up any little blups and boom, the file is out and you are 100% sure that it is ready for your client to drop into their production timeline. Make mistakes with this simple kind of work and you won’t stay in business long.

But much of my work includes vast numbers of files. And if I didn’t have some sort of work flow established and some specialized software, I would not be able to have the confidence that my files are nearly perfect when they leave my studio.

The first hurdle is making sure that you get good clean takes to start with. This takes a keen ear for your own delivery, mouth noises and vowel flutter or other stray noises that creep into the studio. I don’t use headphones anymore when I am self-directing – only when on phone patch or using ISDN – and even then, I try to listen with one ear open. But I can hear that pesky little smacking noise that my mouth makes sometimes when I say a word with an “l” in it.

The second challenge is naming vast numbers of files – sometimes with obscure file names that do not in any way relate to the content of the file in a way that would help you keep track of them. Other times they are at least sequentially numbered – except that the leading zero is left out of a sequence, so the files don’t end up exactly in sequential order according to the brain of the computer doing the sorting. Another time sucker and error prone task is keying in the names of these files one by one.

I am using a new piece of software that helps in these first two steps. Word2WAV lets you record over and over again until you are happy with a take AND saves each previous take as a backup file in another folder. It also automatically names the files with your unique file names or lets you create an accurate sequential numbering system.

While the newest version of W2W includes some editing and punch in capabilities – I find that I switch back to my Adobe Audition for any serious editing that may need to be done. When doing single word files, or short telephony prompts there is not a lot of editing. It is the batch processing that is the next hurdle in the Quality Assurance Process.

Each of my clients has different normalization levels, data and bit rates and file formats, so it is important to keep track of that information so you don’t end up sending out a format or level that is wrong. This not only cuts into your profit, but it puts a crimp in the client’s schedule when you have to redo the work.

I use the batch processing features of Adobe Audition – plus VoxStudio. Both convert sample and bit rates nicely. Audition does a much better job at normalizing a bunch of files. And VoxStudio is great for adding a specified amount of silence to the beginning and end of a file.

Then, after you know the files are clean, in the right format, at the right amplitude and named correctly, you still have to deliver them. Most telephony files are so small in size that it isn’t an issue to send them by email – except that there are so darn many of them. Simple – create a ZIP file and send that. One of my clients wants the files encrypted, so I have a process in place for that as well.

If you are sending wavs or aiffs chances are you will either need to send the files a few at a time through email or use some kind of ftp service. You want to reduce the number of emails going back and forth saying that the client hasn’t received File 23 -34 yet.

And you have to keep backups of what you do. This means some sort of file management system that allows you to quickly find the files you need – if you need to make changes. Clients like it when you are able to make a quick fix easily because you still have the original files close at hand.

Quality Assurance is part of my business plan. And it is something my clients can count on.

Read the Manual

My inclination – when faced with a new piece of software (or pretty much anything) – is to install it and start using it. Read the manual? Not me. There are menus and a help button! (In the case of my cell phone, there is 611.)

Manuals usually have pages and pages of information that I already know and to find the bit of information I don’t know would require formulating a question that the Table of Contents understands.

However, in the case of Word 2 Wav, the manual is simple and short. Had I read the manual, I would have been able to save myself even MORE time this past year. Thank goodness Herve reads the forums.

W2W is a wonderful piece of software that I use at the beginning of my recording process for long form narration and IVR files. Although it is not a finishing program (I use Adobe Audition and Vox Studio for that), it sure saves time in several ways.

Before W2W, I developed a complicated workaround for Vox Studio that included converting a multi-column Excel spreadsheet to a comma delimited txt file, sending it to my Mac, using Text Wrangler to convert it to the format that Vox Studio uses (which separated the first field of information from the rest of the fields), then back to my recording computer, where I added two lines of text to the top and bottom of the file – and then FINALLY record.

After lamenting this time sucking process on a forum, Herve pointed out that W2W does ALL of this in one step (and doesn’t require the addition of the two lines of text).

This is very clearly explained in the manual – and (if I had been a bit more observant), clearly intuitive in the import area of the Setup screen.

Bottom line – there are wonderful tools out there for us to use as we work to meet our client’s needs and improve our ROI.

Lots of mu-law / u-law questions

There has been an uptick in posts to various groups recently about a very standard telephony format called mu-law or u-law.

This is one of the most common telephony formats and there are several tools that can be used to convert to this format – PC and Mac.

If you are on a PC, then most of the audio recording software that professionals are using will do the job – I personally use Adobe Audition.

For Mac, Switch from NCH works well and is pretty easy.

If you are new to telephony – here are a couple of caveats:

1. The sound files use a low sample and bit rate in order to play them back over the phone – they will NOT sound good to your ears when compared to a regular file
2. Always capture at a regular sample and bit rate of at least 44.1 and 16 bit.
3. Downsample to 8K
4. Do NOT change the bit rate to 8 bit or your end result will have a LOT of hiss
5. Use the conversion program to do the final compression to 8 bit

If you are going to be doing a LOT of telephony work, then invest in Vox Studio. It is not cheap, but it will save you so much time, it will pay for itself very quickly.

Adding to the Land Fill

I just spent about an hour and a half going through about 50 pounds of paper scripts that have been piling up over the past couple of years. While I would probably read a lot of these scripts right off the screen, during the school semester I usually bring in scripts from sessions to show my students what a professional voice talent may see in the course of a week.

Scripts range from very “formal” radio and TV scripts (with logos and official titles to help the radio and TV stations figure out which spot to run when), to a hasty email with a single line.

Normally, if I am going to work in my ISDN studio, I need to print out the script, as there is no monitor tied to a computer. I may do that one day (add a monitor), but frankly I like to mark the script when I am in a “live” session with a producer. There are always suggested changes or words that they want inflected a specific way. And there are usually script changes. The same process holds true with most of my phone patch sessions – with the producer right there on the line, it is sometimes better to be able to make quick marks on the script as you work together to get the “right” read.

If I am working self-directed, or using Word2Wav or Vox Studio, I usually read off the screen. If I feel that I need to change an inflection, I do it on the fly.

Other kinds of scripts arrive as pdf attachments in both horizontal and vertical formats – sometimes with font so small it is impossible to read. If it is a straight script with no images, it can be fairly quickly cut and pasted into a text editor and then manipulated. But if the pdf contains a storyboard with many images and a few words per image, it takes quite a bit of time to bump it up to something readable. And it isn’t just that the font is small, it is that the copy is so broken up that finding the continuity is difficult.

Word or Text documents are more easily manipulated to bump up font size and correct widows and orphans – and tighten up broken continuity.  I receive Excel spreadsheets that need to be converted to tables and then to text for importing into my specialized software (mentioned earlier). PowerPoint presentations are also in the mix – which I usually have been reading off the screen. It is nice to see what it is you are talking about by referring to the image above the notes frame, and printing PowerPoint e-Learning presentations do take a lot of paper! However, since these scripts usually require multiple uniquely named files, it would really be better to find a way to get the notes pages into a Word doc for importing into Word2Wav.

After being sure that I had a notation somewhere of the client, the job, the studio (if one was used – either in person or using ISDN) and the producer, agency or production company, I filled up the giant blue recycle bin about halfway.