Last week, Apple introduced iLife ’08, the latest version of their multimedia software bundle – and that means a new version of the impressive GarageBand audio software. Of course, impressive is a strong term, but I would definitely apply it to a piece of software that gives you powerful and easy-to-use recording, mixing, and editing tools as one fifth of an $80 package.

I picked up my copy at the local Apple Store the first day they were available, and I’ve used the new version for about 15 hours of recording and editing in the last week. This review will focus on the new features (both good and bad) I’ve seen in that time, and list a few things that I think were left out.

What’s (Officially) New in Garageband ’08?

Apple’s GarageBand site describes these new features for the software:

  • Magic GarageBand: Play with a hand-picked band on a virtual stage. GarageBand generates a new project based on genre and performance styles.
  • Multi-take recording: Mark a region to repeat, record your part multiple times, and pick your best performance.
  • Automation of tempo effects and instruments: Set multiple edit points in a track to automate EQ and effect changes like a pro.
  • Arrangements: Define sections of your song – intro, verse, chorus – and copy, move, or delete at will.
  • Visual EQ: Graphically adjust frequencies for each track by clicking and dragging individual EQ bands.

Four of these five listed improvements will be useful to many podcasters (while Magic GarageBand has a limited appeal for performers, it offers very little to podcasters), but there are many other enhancements under the hood.

Recording Improvements

Increased Audio Resolution

GarageBand 4 (the official version number of the new software) can now record at up to 24-bit resolution. This means that the software can differentiate between more levels of volume. That doesn’t mean that you can suddenly get louder without clipping, only that the loudness scale is more nuanced than before. Considering that the previous version of GarageBand could differentiate between 65,536 levels of volume (the same range a CD player can reproduce), the increase to 16,777,216 levels is mainly of interest to musicians.

If Apple had added more sampling rates (that’s the 22/44.1/48 kHz number you may have seen), it would be easier to import the audio from a digital video camera into GarageBand. Maybe that will come in the next version.

Record Multiple Takes

MultipleTakesThis will be an interesting feature for some podcasters: the ability to take as many tries at a section as you need, then pick the best one for your final project. This seems particularly useful for re-recording a section of an almost-finished show, as multiple takes on the first time through are as simple as leaving the recorder on and trying something again.

To use the multiple takes feature, click the cycle button in the playback controls section, then move or resize the yellow line that appears in a new track at the top of your project – it should cover the area where you are going to record your takes.

While this is a cool feature when you need it, I would like to recommend it with a warning: In my testing, if I set the yellow selection region so that it overlapped other audio, that audio would disappear from my recording when I finished recording the takes. If you are going to use this, either cut out the section you are going to replace, or just delete it entirely before you start the multiple take process.

Automatic Level Control

Auto Level Control

There is a new checkbox in the track info pane for Automatic Level Control. This feature is similar to the Auto Gain Control you see on some portable recorders, and it seems similarly unwise to use.

What Automatic Level Control attempts to do is adjust the recording input level down to keep loud sounds from clipping and boost it to keep quiet sounds audible. Unfortunately, this means that a few seconds of silence will result in the next sound recorded being much louder than it should be. I did a test of this, and it took a pause of just 5 seconds for the record level to get too loud for a decent recording. In other words, if you check this box, don’t ever pause to go over your notes or to take a drink of water.

New 9,999 Measure Limit

Earlier versions of GarageBand limited projects to 999 (GB 1 & 2) or 1,999 (GB 3) measures, which at the default tempo meant that the longest podcasts you could produce with the software were 33 or 66 minutes. Of course, changing the tempo before you started recording could as much as triple the recording time, but it was still a limitation that frustrated many users.

In GarageBand 4, the limit has been increased to 9,999 measures, giving you 5 hours, 33 minutes and 16 seconds of recording time at the default 120 bpm tempo, or up to 16 hours, 39 minutes and 48 seconds of time if you reduce the tempo to the minimum 40 bpm. If that doesn’t work for you, your podcast is too long!

Editing Improvements


AutomationApple’s list mentions that – in addition to volume automation – you can now control tempo, EQ, and effects with automation curves. What it leaves out is that these automation curves can now be locked to the track: if you move a region of sound, your adjustment curves (volume or otherwise) will move along with it. This one change puts an end to fixing all of your volume adjustments after you realize you missed an “um” and have to go back to edit it out. This is a major frustration I’ve had since I started podcasting with GarageBand two years ago, and it is finally fixed. Enable this feature by going to Control > Lock Automation Curves to Regions.

The advertised automation improvement is of interest as well: Though the new tempo adjustment is mostly something for musicians, the ability to add a bit of echo or EQ to one part of a track can be useful if you want to make a quick fix or enhancement without making a new track and copying the offending section over. For some, this may be a big help, but for me, it is mainly a time-saver for those few instances where you need to do something like that.

Delete and Move

This is a very simple but useful feature for podcasters. It deletes an audio region and moves the regions that follow it earlier on the timeline to take its place. This – combined with the ability to lock adjustments to the audio– will make it far easier to remove a piece of an already-edited podcast without a headache. Also, if you edit by cutting out sections, deleting them, and then moving the later regions toward the beginning of the timeline, this will remove a step from your editing process. Luckily, there is a simple keyboard shortcut for this operation (control-Delete) which gives you an easy choice: “Delete and Move,” or simply “Delete.”

Alignment Guides

Alignment GuidesGarageBand 4 now includes Alignment Guides (Control > Show Alignment Guides), a feature that has been borrowed from Keynote, Pages, and other Apple software. I have a love/hate relationship with Alignment Guides in other programs, and that seems to have carried over to their use in GarageBand as well.

With Alignment Guides enabled, it is far easier to align audio regions and automation tracks with one another. In fact, it not only makes it easier, it makes it almost unavoidable (this is where the love/hate part comes in). When activated, this feature not only conjures little lines to tell you when your regions are aligned, it also snaps the regions into alignment if they are only almost aligned. At the default zoom level, this means that you can’t make adjustments within about 1.5 seconds of an edit or volume automation adjustment without snapping to that nearby point.

This feature is great for making sure there are no gaps in a recording with background noise, but bad when you want to tweak something forward or backward a fraction of a second. Luckily, it is easy to switch on and off when needed.

Arrange Track

ArrangeThe Arrangements feature (Track > Show Arrange Track) isn’t just for musicians to keep the verses and the chorus straight, it can be useful for podcasting as well – but not as useful as I’d hoped.

Arrange allows you to select and move whole sections of a GarageBand Project as blocks – including all of the tracks, automations, and audio regions. Unfortunately, you can’t treat these “Arrange Regions” like you would the audio regions in a project. If you move a region to the left two seconds and let go, it won’t move the contents of that region two seconds earlier in the file. Instead, the Arrange Region will trade places with the previous region in the project – even if that places it 10 minutes earlier in the recording.

I can see the Arrange Track being useful as a way to keep track of the sections of a long podcast, or to rearrange large segments of your show, but it doesn’t behave the way everything else in GarageBand does so I probably won’t use it much.

Type-In Time on the LCD

Time CodeAnother simple one, but a nice addition: if you want to go to a particular spot in your project, you can now double-click on the time display LCD and enter the target time (or measure) to bring the play head – and your view – to that spot in the project.

More Refined Auto-Ducking

I do all of my ducking manually (ducking is reducing the volume of background tracks when foreground tracks come in), but I have heard that the Auto-Ducking feature in GarageBand is popular with some people. Part of the reason I didn’t use it was its lack of options – something that has just been solved in GarageBand 4. The newly-adjustable ducking tool seems to work pretty well when set right, but I still prefer doing it manually.

Visual EQ

Visual EQGarageBand’s equalizer settings have been greatly improved in this latest version, with a new visual EQ editor that includes an overlayed spectrum analyzer. This makes it easy to adjust equalization by picking one of the presets, then fine-tuning the sound by dragging the blue line that represents the adjustments being applied.

This was one of the tools that I didn’t know I needed until I had it – it has proved itself useful many times in the first week I’ve had it.

Exporting Improvements


Garageband has a new checkbox in the Advanced section of the application preferences called Export Projects at Full Loudness. This adjusts the loudest sound in the recording to be at 100% volume, insuring that your listeners won’t have to boost the volume too much on their end to listen to your show.

New Share Options

The Share > Send Song to iTunes menu item now lets you set the metadata (album, artist, etc.) and playlist you would like to use when exporting your song to iTunes. This can save you a step or two when exporting audio that doesn’t fit your default “My Info” settings in the GarageBand Preferences. The other big improvement is the ability to select what format and bitrate your projects are exported in. That’s right, you can now create AIFF, MP3, and AAC files right in GarageBand. In addition, all of these option are available whether you’ve selected “Create New Podcast Episode” or “Create New Music Project” when you first open your file – no more AAC-only podcasts!

In addition to the Send Song to iTunes option, there is a new Share > Export Song to Disk choice as well. This option gives you the same compression and bitrate settings as the iTunes export, then asks you what folder you’d like your finished file deposited in. This is a great feature that allows you to bypass iTunes entirely – as long as you don’t use it to add artwork or shownotes to the file.


GarageBand ’08 (or GarageBand 4, whichever name you want to call it) includes many time-saving and headache-saving improvements over the previous version. It also includes some new features that – if used properly – will improve the overall sound and quality of your podcast. For a heavy GarageBand user, I see these improvements as nearly worth the $80 pricetag for the iLife suite. Of course, iLife includes iPhoto, iMovie, iDVD, and iWeb in addition to GarageBand, so the purchase should be a no-brainer for the Mac-based podcaster.


A while ago, I wrote a post where I said that using Skype for interviews was a bad idea. I still stand by the premise of that piece – that Skype-to-Skype connections should not be used in most situations – but a few things have happened in the last few months to change my mind about the blanket statement “Skype is bad for recording interviews.”

The first thing was pretty simple. The telephone hybrid interface I had been using stopped working. This box took the sound from my studio mic (through a mixer) and fed it into the phone line so my interview subject could hear me. At the same time, it fed just the caller’s audio into a different channel of the mixer – allowing it to be recorded independently from my voice. These devices are usually very good at what the do (they are what professional radio producers use to put phone calls on the air), but I was suddenly without mine, and I needed to decide if it was worth it to get another one.Skype Unlimited Icon

At about the same time, Skype announced a new option for their SkypeOut service: Skype Unlimited. For $30 a year, you can get unlimited calling to anyone in the US and Canada. That’s unlimited calling to real phones – in other words, an inexpensive digital connection between your computer and your interviewees’ in-home recording systems (AKA their telephones).

These two developments prodded me to look into Skype as a way to record phone calls, and I was impressed with what I found: with some inexpensive third-party software, I get better sound quality using SkypeOut than I did with expensive dedicated hardware.

Call Recorder WindowRecording Software

Since I use a Macintosh system for my recording and editing, I downloaded a trial copy of Ecamm Network’s Call Recorder ($15, includes lifetime upgrades) to test out. Call recorder adds a small recording window (complete with level meters) to Skype, and is very easy to use. It adds a “Recording” tab to your Skype preferences, giving you full control over recording quality and compression (including an uncompressed option).

In my opinion, Call Recorder’s killer feature is the ability to have each side of the phone call output as an independent AIFF file. If you edit your interviews (or even apply any processing to them after the fact), this is a tremendously useful feature – it’s one of the main reasons people pay hundreds of dollars for a telephone hybrid. Because the whole phone call happens inside your computer, there is complete separation between the two sides of the call, allowing you to edit the interview as two separate recordings that can be spliced together into one smooth final product.

I don’t use Windows very much, but I did find a piece of software that seems to duplicate Call Recorder’s functionality for that platform: it’s called CallBurner ($50, trial version available), and what I have seen written about it so far seems good.

What’s the Catch?

There isn’t much of a catch, as long as you have a moderately fast, reliable internet connection. However, this could be harder to achieve than you think. When I started this experiment, I tried calling over a cable internet connection and (quite surprisingly) the results were terrible! After some research and experimentation on my part, I figured out that the connection was fast, but it was only fast in little spurts. Even though the average throughput was high, the speed dropped almost all of the way to zero many times a second. [I’ve gotten a number of question about this, so I want to emphasize that this was just my specific experience. Most people have no problems whatsoever using Skype with a cable connection.]

When I tried another test on a T3 connection, the results were much better. Don’t let that “T3” speed scare you, though – switching to DSL at home fixed my connection problems even though the speed is lower overall.

So You Want to Do a Phone Interview…

Now that we have this inexpensive tool for recording phone calls, it is far easier for individuals to capture famous or faraway voices for their podcasts or other audio projects. If you want to take advantage of this new access, here are a few tips:

  • Don’t use a cheap mic on your end.
    Use a good USB mic, or plug your mixer into the computer to get your sound as good as possible.
  • Always call a “traditional” landline unless cellular or VOIP is the only option.
    Cell phones are inherently flaky, and while VOIP calls are pretty reliable, there is the potential for some digital weirdness there – it’s better to divide the chance of this by two by only doing the VOIP thing on your side where it is necessary for the recording.
  • Don’t let the guest use a speakerphone or cheap headset.
    You can tell when someone is using one in a normal phone conversation, so your listeners will be able to tell as well.
  • Ask your interviewee to close the door, turn off their cell phone, and give you their undivided attention.
    You’ll be able to hear rustling papers, typing, and office noise in the recording – just like in a regular phone call.
  • Treat it like an interview, not a phone call.
    The phone is your recording system, not an excuse to have a casual chat (unless that’s what you are looking to record). If you need help with interviewing, read my Interview Basics post to get you started.

If you want a sample of what a Skype-to-phone recording sounds like, listen to the segment below.

icon for podpress  SkypeOut Interview Sample [12:00m]: Play Now | Play in Popup | Download

Microphone Types

Print This Post Print This Post

There are many online resources available that can tell you all about the nuances of microphones, and I’m not going to try to reproduce them here. Instead, I want to give you a quick primer on a few different types of microphones and what they can be used for.

Omnidirectional Microphones

These do pretty much what they say they do: record sound from all directions. They will pick up more from the front than the back, but they give you a good dose of all of the sound in the area.

Omnidirectional mics are great for getting the feel for a place. They will record ambient sounds without emphasizing any of them too much (besides the fact that physics dictates that nearby sounds will be way louder than far away sounds). In an interview, an omni mic will give a sense that you are in a location, but can pick up background sounds in a distracting way if you aren’t careful. Put one close enough to the subject and you’ll get a good mix of foreground and background sound — provided that the setting isn’t too loud.

Cardioid Microphones

These are more directional than omni mics and are sometimes called unidirectional microphones. They come in different “amounts of directionality,” from slightly directional to very directional. They also pick up very well from the back of the mic — where the handle of a handheld mic usually is. This means that some cardioid mics are very susceptible to handling noise and vibration.

Using a cardioid mic gives you that “radio sound” you are used to from DJs and talk radio hosts. This is because of something called the Proximity Effect that enhances the bass in a voice when the microphone is very close to the speaker’s mouth (close as in a few inches away). The main reason for using a cardioid mic in a podcast is to reduce the background environmental noise. A directional mic will not eliminate background noise, but it will make it less distracting.

Hypercardiod Microphones

These can also be called shotgun microphones, depending upon their characteristics and shape. Hypercardiod microphones pick up sound from a very specific area in front of the mic, and can share handling-noise issues with their less-directional cardiod siblings.

These mics are not magical, so don’t expect to use one to record a speaker 20 feet away and get amazing sound. For interviews and other spoken-word recordings, you should treat these microphones just like the others. They will allow you to record in situations that would be hard with other mics, and they can give you a studio-like sound in the field, but they don’t work miracles.

Stereo vs. Mono

The difference here should be pretty obvious: In stereo, the left and right speakers produce different sounds, and in mono they produce the same sound. When it comes to microphones, you need two to record stereo and one to record mono. Of course, there are many microphones that include two recording elements in one casing, so stereo doesn’t always mean two wires running to two devices.

Unless you are podcasting music, stereo isn’t a necessity for most shows. In fact, getting really close to a stereo mic can cause some very weird issues with your sound. However, an exception to the “stereo is unnecessary” rule can be seen in my Hat Mics post, where I use a pair of stereo mics to record two separate tracks into one file.

Specialty Mics

There are many types of microphones designed for very specific tasks. I won’t go into the details of all of them but here are a couple of types that could be of interest to podcasters:

XY Stereo Microphones: For recording a more directional, authentic-sounding stereo sound. These are some of the only stereo mics that can easily be used for voice recording like a mono mic without sounding strange. (See the built-in mics of the Zoom H4 for an example of an XY setup.)

In- Ear Binaural Microphones: Because they sit in your ears, mics like these from Sound Professionals give a 3D feel by recording exactly what you hear. The catch is that they only give that 3D feel through headphones, otherwise they are just stereo mics. You can hear a recording I made with this type of microphone below.

icon for podpress  BBTS5HaymarketExtra [4:32m]: Play Now | Play in Popup | Download
Page 1 of 5 12345
© 2010 Adam Weiss: Podcast Consultant