Continuo Admin

1. Your voice not set

Record about 60–90 seconds of natural speech (quiet room, normal pace). Reading the passage below works well. This creates an ElevenLabs instant clone — re-recording replaces the previous clone.

“Hello, my name is … and I’m recording this so my patients can hear information about their condition in my own voice. Clear information helps people feel confident about their treatment: what their medicine is for, how to take it properly, and what to look out for while they recover. If anything in a video is unclear, or symptoms change, patients should always contact the practice for advice rather than waiting.”

2. Your selfie not set

A well-lit, front-facing head-and-shoulders photo on a plain background. This becomes the HeyGen talking photo — uploading again replaces it.

3. Generate a video

Pick a condition, tweak the script, generate. The pipeline is: script → your cloned voice (ElevenLabs) → talking-photo render (HeyGen) → hosted on this worker. Renders take a few minutes. The voice sliders also apply to the test box in section 1 — tune there before spending render credits. Punctuation shapes delivery too: shorter sentences, commas and dashes add natural rhythm, and <break time="0.6s" /> in the script inserts a pause.

Expressiveness Style exaggeration Speaking speed Engine (HeyGen) Head motion (Avatar III only) Expression (HeyGen) Background colour

4. Videos

Links are unguessable but public — fine for testing, not for patient data.

Continuo Admin

Who are you?

Continuo Admin

1. Your voice not set

2. Your selfie not set

3. Generate a video

4. Videos