Avatar
A talking face for your assistant: give it a portrait and the text to say, and it streams a video of that face speaking.
Overview
Agentyk Avatar turns a single portrait into a speaking video in real time. It can speak your text with Agentyk Voice, or move along with audio you send yourself. The video arrives in short pieces while it speaks, so a conversation feels live.
Agentyk Avatar
On requestGives your assistant a face that speaks.
A voice on its own works well on the phone, but on a screen people like to see who they are talking to. Recording a person for every new answer is not an option when the answers change all day.
Send Agentyk Avatar one portrait and the text to say. It speaks the text with Agentyk Voice and streams back a video of that face talking, in pieces of about a second, so playback starts while the rest is still being made. If you have your own audio, send that instead and the face moves along with it. When the user interrupts, your app stops it at once and moves on.
- One portrait in, a talking video out
- Speaks your text with Agentyk Voice, or follows your own audio
- Streams the video while it speaks, about a second per piece
- Square video from 256 up to 512 pixels
- Stops at once when someone interrupts
- Sessions of up to 30 minutes
- Available on request: ask us for access, and use a portrait you have the rights to