Reallusion Crazy Talk Animator Pro review

Crazy Talk Animator Pro aims to be an easy way to create 3D animations. The software comes with a selection of 2D cartoon characters, but the main focus is on creating your own animated avatars – what the software calls ‘actors’- from photos.

You can create an actor’s head and body from a single shot, but we had better results importing separate photos for each – the software defaults to cartoon-like large heads and small bodies, and you’ll be able to create more effective facial animations later with a more detailed head shot. The first stage is cropping out the head and doing some basic level adjustments. You then outline the basic facial shape by defining the edges of the eyes and mouth, followed by advanced editing. This involves adjusting numerous control points, to define the shape of the face, eyes, eyebrows, nose and mouth, as well as eyes and teeth. So the software can animate the eyes and teeth, you replace those from your photo with computerised versions – this is the most fiddly part of face creation, as it’s hard not to create someone who looks like an animé character or Esther Rantzen.

Results when importing a body from a photo were more mixed. For a start, it takes a long time to cut the body out from its background. Instead of selecting the part you want to cut out and keep, you have to use a masking tool to erase the parts of the picture you don’t want to use, which involves a lot of mouse work. You also have a lot less fine control over defining the parts of the body, so your final actor can have spindly arms and jagged edges on his legs. Once we found the right pose for the photo – arms out to the sides, feet at 10 to two – we managed to import a passable computer-generated version of a human body. There was one major frustration in that you can’t go back through the creation steps, so we often had to start actor creation from scratch when something didn’t look right at a later stage. The program also crashed a couple of times, and as there’s no auto-recovery we lost our creations. We’d recommend saving your project often.

Once you have a character, you can start animating. Most of this is done in real time. You insert actors, props and backgrounds (these can come from the built-in library or any image you choose) then import music, sound effects or recorded dialogue and select from a list of preinstalled animations for head and body – anything from walking to dance moves to hand gestures. The software continues recording as you select each preset, so if you’re quick enough you can match gestures and movements to events on the soundtrack. We sometimes wished the software wasn’t so ready to start recording every time you clicked on something – while you can edit your animation later in the timeline editor, we sometimes got fed up of deleting animations when all we wanted was a preview.

For more fine control over your actor, you can use puppet mode. This provides you with a selection of base animations for the head and body – such as frown or wave hands – and you control them by moving the mouse around as you record. Puppet mode has a half-speed setting, so you have more of a chance to match your movements to the soundtrack, and you can move animations around in the timeline later. Once you’re finished you can export the animation to a movie file, but the program crashed when exporting XviD so we had to use WMV.

Despite having almost no experience of animation, we found it relatively easy to import a model and create a simple animation using presets and puppet mode. We wish there was more control over real-time recording so we had to fix less in the timeline editor and the crashes were annoying, but this is still an incredibly simple way for beginners to animate.