Can Heygen synchronize external audio without generating its own content?

Hi everyone, I have a question about Heygen. Currently, we have all our response logic based on a fine-tuned LLM, and we generate our answer's audio using LiveKit. My question is: is it possible to use Heygen solely to feed it the resulting audio so that it produces a video that simply repeats what the audio says? In other words, we don’t want Heygen to generate the content itself, but rather just synchronize the audio we provide with an avatar or something similar. Thanks in advance for any insights.