Audio is a critical part of an animatronic show. Obviously you need technically high quality sound (no noise, intelligible speech, etc.), but there is much more to it. Because the motions of the characters are fairly limited, you will be using the audio to convey much of the emotion of the show.
In the past, audio would be recorded on magnetic tape. But now, we generally record directly to a computer. The software that allows us to record and manipulate sounds is called an audio editor. Audacity is an audio editor that runs on both Windows PCs and Macintoshes, and is freely available on the web at:
Audacity is a very powerful tool with many advanced capabilities. For complete information on Audacity, see the manual and the website.
There are lots of other good audio editors you can use that aren’t free. You can find a large list here:
If you happen to have access to a Macintosh, GarageBand is a good choice. It includes lots of little audio “loops” (short recordings) including various jingles and sound effects. Not only are these very useful, they are “royalty free” – in other words, you can legally use them without having to ask permission.
The goal is to create the audio files needed by Visual Show Automation, or VSA for short. VSA includes a tool which analyzes an audio file and generates corresponding servo commands to move a mouth in sync. For this to work as intended, you need a separate audio file for each character (no background music or effects) for the analysis. In addition, you will also need the a version that includes everything (all characters, sound effects and background music) that plays during the performance.
Most audio editors support “multi-track” editing. This means that you can record each character independently as a separate “track”. Sound effects and background music can also be added on separate tracks. You can then create “mixes” which add these together in various proportions. This is the way most professional studio recordings are done. For example, the voices for most animated films are recorded one person at a time – the other actors aren’t even there. All of these separate recordings are then mixed together to create the audio the audience hears. In some cases, the actors having a conversation in the film have never even met!
Here are some general tips for recording:
1. Find a quiet place to do the recording.
2. Inexpensive PCs often have very noisy power supplies which leak into the audio input. I have found that using an external USB audio input helps quite a bit since they tend to have good power filters. I have found that the inexpensive Logitech USB Desktop Microphone works fairly well. If the voice actor is fidgety, a headset may work better. Often young actors will lean in and out from a desktop mic causing the volume levels to modulate weirdly.
3. Watch your volume levels. Recordings that are too quiet will suffer from noise. Ones that are too loud will distort. It generally helps to keep your voice talent a fixed distance from the mic and ask them to speak at a fairly constant volume.
4. Don’t be afraid to use the tools available in your editor. For example, most editors include noise gating or some other type of noise reduction. This can make a dramatic difference. You can also manually chop out quiet segments so that they are truly silent. Pitch shifting a voice is great for making a voice that may be more in character. Little characters may have higher-pitched voices than any of your talent!