Open AI Talking Avatar on AIAI, add your image and voice track, then describe the presentation you want without submitting a generation.
This guide prepares AIAI AI Talking Avatar with one existing product image, one short owned voice file and a manual prompt. It does not create or evaluate an avatar.
- Beginner
- English video + illustrated steps
- Observed September 19, 2026
- 0 generation jobs
Disclosure: eWiki has a business relationship with AIAI. See our disclosure. Interface labels and quotes can change.
Watch the walkthrough
Real continuous window recording with reviewed waiting intervals accelerated. English narration uses en-US-JennyNeural. Zooms and pointer guides are editorial annotations, not recorded cursor telemetry.
1. Check model and mode
Our session kept Kling Avatar V2 and Standard mode. Pro mode was visible but not selected or tested. Model changes can alter fields, limits and cost.
2. Upload image and audio
Prepare a still image and a voice track you own or are permitted to use. We uploaded the existing product-reference.webp and setup-sample.mp3. The audio was about six seconds and 37.7 KB; it was uploaded but not played.
3. Write the presentation prompt
The prompt asks for a calm neutral presentation, gentle head-level movement, steady studio lighting, and no added text or scene cuts. Write your own actions and camera language; Prompt Assistant was not used.
4. Review the completed setup
Public Visibility was on and was not changed. The final button displayed Generate 49, up from the initial 14 after the inputs were added. This is a quote, not a charge. We did not click Generate Now or Generate.
Open AI Talking Avatar on AIAI with your own image and voice. Review permissions and the quote before any future generation.
Before you continue
| Question | What to check |
|---|---|
| Can I use any face or voice? | Use material you own or have permission to use, and follow the platform’s current policy. |
| My upload fails | Check type, size and the visible upload message. Do not generate to test an upload. |
| The quote changes | Quotes can depend on mode, inputs, plan or product changes. |
| How realistic is the result? | Unknown here. No avatar was generated or reviewed. |
Session notes
Recorded in signed-in desktop Chrome on September 19, 2026. Actions: model and mode review, existing image upload, existing MP3 upload, manual prompt entry. Generation count 0; credits consumed 0; displayed balance 6 → 6. No assistant request, translation, audio playback, payment, result generation or download was submitted.
Improve this guide
Report a problem or suggest the next question
Submissions enter a private editorial queue. They are reviewed and deduplicated before any public update or new page is created.