Course Workshop & Avatars
My avatar
Lecturers make themselves the speaker of their courses: consent, photo or video selfie, voice, test, use — revocable at any time.
Varför#
Your own course with your own face and your own voice — without a studio and without an appointment.
Så fungerar det#
KlickvägLearning workshopMy avatar
- 1Open “Learning workshop” → “My avatar” (address /werkstatt/ich). At the top the four steps: photo, voice, test, use (Fig. 1).
Fig. 1 The four steps to your own avatar: photo, voice, test, use. - 2“Your consent”: “Name as it appears in the course”, tick “I am the person depicted …” and “I am aware: my avatar will be … labelled as AI-generated”, optionally “My voice may also be cloned” — “Consent and continue” (Fig. 2).
Fig. 2 “Your consent”: name in the course, the two confirmations and the choice of whether the voice may also be cloned. - 3Photo: “Record video selfie” (four seconds, the sharpest frame is taken) or “Upload photo” — straight into the camera, mouth closed, even light from the front, head and shoulders in the picture. Then “Use this image”.
- 4Voice: “Own voice (30 s)” — read the displayed text calmly, “Finished speaking”; or choose a ready-made voice.
- 5Test: “Listen to test sentence” — the moving image is created with the first course video.
- 6Use: “Use as speaker” — afterwards the avatar can be chosen in all your own courses when placing the speaker.
Inställningar och reglage#
- Kan slås på per kundStandard: påModul: Eigene Avatare
- Part of the module “Own avatars” (requires the module “Avatar lecturer”).
What the assistant checks#
- No person clearly recognisable → “Please move closer to the camera.”
- Low camera resolution → the image is upscaled slightly; a photo from the phone will be sharper.
- Recording too short or almost silent → once more, completely and closer to the microphone.
- Camera or microphone not permitted → notice to allow access, or upload a photo.
Image, voice and labelling#
Photo, voice sample and voice capsule are kept in the platform's storage in Frankfurt. Image and own voice are computed by the platform with its own models on rented graphics servers. If a preset voice is chosen instead of one's own, the text to be spoken goes to a specialised speech service in the USA — only this text. Consent can be revoked at any time. Everything the avatar says is labelled as AI-generated for learners.
The rule “Every face speaks only with its own voice” applies: without consent to your own voice, the avatar speaks with a chosen ready-made voice — never with another person's voice.
Who may do what#
Lecturers with teaching authorisation, workshop authors, administration and operator — each only for themselves.
Tips#
- The green background during building is intentional; it is removed afterwards so that the figure stands freely on the slides.
- You can close the page; the build continues, and the status is shown there on your next visit.
- “Withdraw consent”: the avatar then stops speaking immediately — in no course and not live.
- The figure only appears in your own courses, not with colleagues.
Relaterade funktioner#
- Avatars & voicesYour own figures from a photo and your own voices from a recording — with consent as a condition and one fixed rule: every face speaks only with its own voice.
- Phase 3 · Speakers & InteractionWho speaks the course, in which languages, with which voice — and where the speaker stands on every slide.
- Speaker spotDecide slide by slide where and how large the speaker stands — and whether at all. The calculation makes a suggestion, the author approves.