From one portrait to one speaking clip
Turn one authorized portrait into a 720p talking video. Add spoken text with a built-in or saved voice, or supply audio you are allowed to use, then review the complete result in Studio.
Make a photo talkOne portrait · one result · 66 credits shown before submission
The source is one still face. The spoken words or authorized audio supply the delivery. Studio returns one new video that you can play from beginning to end before deciding whether it is ready to share.
A single adult face gives the video its identity and framing.
A short script or audio track supplies the words, pace, and pauses.
Review mouth motion, eyes, facial edges, and audio as one clip.
A poster only proves that the image loaded. Play the clip to inspect the moments that matter: the first open mouth, faster syllables, blinks, skin around the lips, and the final frame.
Good input is practical, not glamorous. Use one adult facing the camera with enough face detail to review, a mouth that is fully visible, simple lighting, and some room around the shoulders.
Use the script path when you need Studio to create the spoken delivery. Use an audio file when the exact recording is already part of the approved message. Both paths still need a close review of the finished video.
SCRIPT PATH
Enter only what should be spoken. Keep production direction out of the line so the message has a natural rhythm and a clear close.
“Welcome. Here is the next step. Take a moment to review it before you continue.”
AUTHORIZED AUDIO PATH
Upload a clean MP3, WAV, or M4A when the wording, pace, and voice are already approved. This demonstration uses a project-owned audio track as the input.
The public page explains the decision. Uploading, the 66-credit quote, consent, task status, result preview, and download all happen in Studio.
Choose a JPG, PNG, or WebP portrait you are authorized to animate.
Write a short script with a voice choice, or attach audio you are authorized to use.
Studio displays 66 credits and asks you to confirm rights before the task starts.
Play the 720p output, then download only after a complete check.
Talking photos work best when the message sounds like a person speaking it. Short sentences, punctuation that creates a pause, and an ending with a little room make the delivery easier to review.
Spell out unusual pronunciations when necessary, separate numbers that could be read more than one way, and place a period where you would want a real pause. Save visual direction for a different field or a separate note.
Do:
One message, one supporting detail, one close.
Avoid:
Dense paragraphs, claim-heavy scripts, and last-second tongue twisters.
The last step is not a thumbnail check. Play the whole result once with sound, then revisit the moment where the message starts, the fastest phrase, and the final pause.
Use a talking photo when a static identity is already right, but the next message needs a face and a voice. The choice is about a focused speaking moment, not replacing a full video production.
Give a clear creator portrait one concise product explanation, then review every statement before it reaches an audience.
Turn an approved educator or host portrait into a short welcome that points learners toward the next step.
Use a direct message with an explicit date, time, and close when the audience needs an update rather than a new scene.
Talking tasks look similar after they are finished. The important difference is the starting material, because that determines what the workflow is allowed to preserve or create.
A talking photo cannot guarantee a perfect result from every source. It is the wrong first choice when the face is hidden, the image contains multiple people, the original video already has motion worth keeping, or the task really needs a new scene.
Choose a clearer source before you start.
Use a single adult portrait for a focused result.
Use Lip Sync when existing motion must remain.
Do not expect the workflow to restore an unusable face.
Keep the decision grounded in the live Studio workflow rather than a generic video promise.
Questions before you continue
These answers describe the current workflow and its limits. They do not promise identical outputs or automatic publishing.
Each route begins with a different source, so use the page that matches the material you already have.
Open Studio when a single portrait, a focused message, and one reviewable 720p talking result are the right next step.
