Voice cloning
Producing a synthesized voice that sounds like a specific person from a sample of their real voice. Used in Ajeris primarily for the voice receptionist (the business owner's voice answering business calls).
Why it matters
A voice receptionist that sounds generic is a chatbot in voice form. A voice receptionist that sounds like the business owner is genuinely close to having an additional staff member who is always available. The technology is the difference between novelty and product.
Voice cloning is also ethically charged. Done legitimately (cloning your own voice with explicit consent), it is a useful tool. Done illegitimately (cloning someone else without consent), it is a real harm. Ajeris explicitly forbids the latter; setup includes a recorded consent step.
How Ajeris handles this
Two clone modes. Instant clone: 30-second sample, ~30 seconds to produce a usable clone. Good for most operational calls. Long-form clone: 5-10 minute sample, multi-step training, ~24 hours to complete. Studio-quality output. The cloned voice is bound to the user's account; deletion is one command and removes the model within the documented retention window.
