manyframeWan 2.2 Speech to Video.
A permitted visual and a prepared voice track become the starting point for a talking shot. Let the spoken script set the timing.
Available models and settings are shown in your studio. Review the quote before generating.
The route.
The right inputs.
Start from the creative job, then choose the model and settings that fit it.
Model ID wan2.2-s2v
- Resolution
- 480P or 720P; default 480P.
- Timing
- Output follows the audio; independent clip duration is not forwarded.
- Optional duration input
- If supplied, must be positive and under 20 seconds; it does not set output length.
- Media boundary
- Direct image and audio inputs, rather than a generic reference collection.
Capabilities describe the Manyframe connector. Your studio shows the currently enabled route, supported settings and maximum charge.
Write the voice track before the shot.
Keep the script short, clear and comfortable to say aloud. Prepare a voice track with a clean opening and ending, then choose an image that suits the delivery. The audio provides the timing, so adding an arbitrary video length will not reshape the spoken performance.
Use only imagery, likenesses and voices you have permission to use. Check mouth movement, facial stability and the relationship between speech and image before presenting the result as a spokesperson or character.
For a generated short narration, review Qwen3 TTS Instruct Flash separately and confirm the available voice. For a silent product move instead, use Wan 2.6 I2V.
Good questions. Clear answers.
Still curious? Browse all FAQs
Does this model create the voice too?
This connector requires supplied audio and an image. Generate or record the voice track separately, then use it as the speech input.
Can I request a fixed five-second output?
The output follows the audio. A supplied duration is checked as positive and under 20 seconds, then removed rather than used as a timing control.
Does connector support mean a UGC offer is ready?
No. The route is implemented, while current availability, consent requirements, voice/likeness permissions and output quality must be assessed for your intended use.
Give the script a visual presence.
Available models and settings are shown in your studio. Review the voice, image permissions and quote before generating.