A face introduces your AI influencer. A voice can explain an idea, welcome an audience or give a product demonstration a more personal rhythm. CamTalk is a coming-soon video feature being designed to bring those choices together in FanPro Studio.
This preview walks through the planned workflow in the supplied designs: choose a character image, prepare the speech and audio, then describe how the character should perform. It also explains the difference between using an influencer’s assigned voice and creating a freeform clip with another voice or uploaded audio.
What CamTalk is being designed to do
The designs place CamTalk inside Videos. Instead of putting everything into one prompt, they separate the character image, spoken words, voice and background sound from the scene direction. That gives each part of a talking video a clear role.
| Input | What it contributes |
|---|---|
| Character image | The visible person and the starting appearance. |
| Speech or uploaded audio | The words and delivery the viewer hears. |
| Voice | The sound of the speaker, using the selected influencer’s voice or a freeform choice. |
| Background sound | Audio that supports the setting. |
| Prompt | Scene direction: expression, gestures and the camera. |
| Optional preset | A starting point that can fill in a prompt and associated settings. |
The useful starting point is a short idea with one purpose. An introduction, a product explanation or a single answer to a common question gives you a focused first clip to review.
Start with a clear character image
The planned Character Image field asks for the full face to be visible. Choose a reference where the expression, face and framing are easy to read. If your intended scene includes hand gestures, leave enough room in the composition for those actions.
The image picker includes My Generations and Uploads. The selected-influencer design filters the library to that influencer, making it easier to find a suitable reference without mixing identities. The designs also provide a freeform path without a selected influencer.
- Decide which character the clip is for.
- Choose a clear image with the full face visible.
- Check whether the framing supports your intended gesture and camera direction.
- Keep the outfit and setting consistent with the story you want the character to tell.
Understand the two planned voice paths
| Path | Planned behavior |
|---|---|
| An influencer is selected | Their assigned voice is used automatically. The design asks you to deselect the influencer before choosing another voice or uploading speech audio. |
| No influencer is selected | The freeform flow lets you choose a library voice for typed speech or supply your own audio. |
| Uploaded speech is active | Remove the uploaded audio before returning to a library voice and a typed script. |
These choices answer different creative needs. An assigned voice can help a recurring influencer sound consistent from one clip to the next. The freeform path gives you room to choose a different voice for a separate concept or work from an audio recording you already have.
The planned voice library includes preview playback and filters for qualities such as style, tone, age and language. Listen for whether the voice fits the character and the message. An energetic opening and a calm explanation may need different delivery even when they share the same subject.
The Upload Audio designs list MP3, WAV and FLAC. Final file limits and supported options will be confirmed with the release. The preview does not establish a voice-cloning feature.
Write the speech and scene direction separately
The speech field is for the words the character should say. The Prompt field is for what the viewer should see: expression, movement and camera direction. Keeping them separate makes the brief easier to read and revise.
These are illustrative writing examples, not tested CamTalk prompts. Notice that the speech explains the subject while the direction describes a performance. A stage instruction such as “smile and turn toward the camera” belongs in the direction, rather than being mixed into words that are meant to be spoken.
- Write one clear opening sentence.
- Give the clip one main point before adding supporting detail.
- Read the script aloud and leave space for pauses.
- Describe a small number of gestures that support the words.
- Check that the reference image can support the planned action.
Add background sound or start from a preset
Background Sound is a separate part of Audio Setup in the designs. Think about what the setting needs: quiet room tone, an outdoor atmosphere or another appropriate background. Supporting audio should leave the speech easy to understand.
The optional CamTalk to Copy input opens CamTalk Presets. The planned library separates Presets from My Templates. Selecting a preset can fill the prompt and associated settings, which you can then review and edit for your own character and script.
Treat a preset as a starting point. A useful template still needs to fit the image, words and intended audience. Check any filled direction for references to props, gestures or a setting that do not belong in your version.
Ideas to prepare before CamTalk launches
| Use case | What to prepare | What the clip should communicate |
|---|---|---|
| Influencer introduction | A clear portrait and a short welcome. | Who the character is and what the audience can expect. |
| Product explainer | A suitable character image and accurate product details. | One feature or detail in straightforward language. |
| Social video opening | A concise hook and one follow-up point. | Why the next few seconds are worth watching. |
| Guide introduction | The goal of the guide and its starting context. | What the viewer will learn before the instructions begin. |
Prepare the material you control now: the character reference, script, intended audience and any audio you have permission to use. A small, organized brief will be easier to adapt when the final controls are available.
The designs include duration, resolution, generation and result states. Their displayed numbers are mockup values, so this preview does not announce final prices, credit charges, clip-length limits or output resolutions.
What to check when the feature becomes available
Once you can generate a CamTalk clip, review the whole performance. A talking video needs the image, sound and timing to work together. A convincing first frame alone will not show whether the complete delivery fits the brief.
- Check that the character remains recognizable throughout the clip.
- Compare mouth movement and spoken audio for timing.
- Listen for clear wording, appropriate pronunciation and natural pauses.
- Watch whether expressions and gestures support the message.
- Check that background sound leaves the voice understandable.
- Review the first and last frames and the actual output format.
Change one part of the brief at a time when refining a result. If the delivery feels rushed, simplify the script before adding more actions. If the gesture distracts from the message, reduce the movement while keeping the words steady.
Explore the video workflows available today
CamTalk is coming soon. While its workflow is being developed, you can explore the existing Studio Marketing examples below to practice organizing character references, dialogue and scene direction.