動画
Talking head
A talking head is a shot of a person speaking to the camera. In AI video, it is a face animated by speech: from one picture and a voice, a model moves the lips, jaw, eyes and head so the face appears to say the words.
Most talking-head tools render ahead of time: you give a script or an audio file, and a clip comes back a while later. A real-time talking head renders as fast as the voice plays, a few frames at a time, which is what makes an AI video call possible: the face answers while you are still in the conversation.
The picture sets everything the voice does not: the face, the framing, the light. A clear, front-facing portrait gives the steadiest result; a full-body or back view leaves little face to animate.
eroqでは
eroq Live renders a real-time talking head of your character on a video call: the picture on their card, 512×512 at 25 frames a second, lip-synced to their own voice as they answer you, 100 credits a minute. It is a conversation, not a clip: eroq has no tool that lip-syncs a video to an audio track you supply.
プライバシー
プライベート、そして制限なし
- →eroq独自のエンジンにフィルターはなく、明文化されたポリシーがひとつあるだけ。
- →何も保持せず、あなたの作品を学習に使うこともありません。
- →プライベートモードはファイルを一切残しません。
よくある質問
What is a talking head video?
A video of a person, or an AI face, speaking to the camera: a presenter, an explainer, a testimonial. AI talking heads animate a portrait from speech instead of filming someone.
Can a talking head be real time?
Yes, when the model renders faster than the voice plays. On eroq Live a character's face is rendered as they speak, 25 frames a second, while you talk with them.
Can I make a talking head video from a script on eroq?
Not as a lip-synced clip: Live answers in conversation, in the character's own words. For a scripted line, render the voice in the voice studio and lay it under the picture in the Editor, without matching lips.