You could use Audition to start/stop pre-recorded audio with kb click. Character Animator will respond if the audio devices are set appropriately.
On Window 10 there is the Narrator feature, which converts text to audio in real time. I've never really used it, but it might be worth a look. I believe there is something similar on Mac.
Nothing like this inherently inside Character yet, but triggered audio is a common request and one we'd love to include in future versions.
Audition has a text to speech feature, so if he can type, the computer can synthesize speech from the text:
On the Mac, I know there are several ways to pipe audio from one application to another(soundflower, loopback, etc,). This would allow that speech to control a character's mouth.