Keep the first script short
A single message is easier to review. Say names, numbers and product terms aloud in the voice preview before paying for video generation.
AI AVATAR / SHOTTIDE
An AI avatar, with your words.
Start with a portrait. Write a short script or bring your own audio, listen to the voice, then turn the image into a talking video.
Choose the right starting point
A talking avatar animates a still portrait using a voice track. ShotTide keeps the portrait, narration and video as separate saved assets, so you can change a line or try a different delivery without rebuilding the entire project.
How to create a talking avatar
- 01
Choose a portrait
Upload a PNG, JPEG or WebP, choose a template or saved image, or design a new character. Use a clearly visible face with enough space around the mouth.
- 02
Prepare and review the voice
Type your script, choose a voice and language, and adjust speed, pitch, emotion or pauses. Add exact text substitutions for tricky names and preview the spoken wording, or import recorded audio. Listen before marking it reviewed.
- 03
Approve the video quote
The studio measures the audio and shows the credit cost. Add a short motion direction, generate the video, then preview and download the saved result.
What to check before you publish
Inspect face and motion
Check mouth shapes, eyes and the outline of the face in the downloaded video. A generated portrait and voice do not guarantee identity consistency or precise lip synchronization.
Inputs, limits and credit costs
- Portrait
- PNG, JPEG or WebP; up to 5 MB and 16 MP.
- Talking video
- Reviewed audio up to 180 seconds in one Kling Avatar V2 request. Prepared audio is compressed below the provider limit.
- Cost
- Voice: 0.2 credits per started 100 characters. Video: 3 credits per started second, with a two-second minimum.
- Captions
- After generation, transcribe the video, review text and timing, choose a style and export a captioned copy. Transcription is charged separately.
Frequently asked questions
Can I use my own photo and voice?
Yes. Upload a portrait and import MP3, WAV or M4A audio, or generate narration from a script. Use a photo and voice you have permission to use.
Does changing the script start another video?
No. Create and review the new audio first. Video generation starts only when you approve its quote and press the generation button.
Can I replace speech in an existing video?
Use AI Lip Sync for an existing MP4, MOV or WebM. This AI Avatar Generator starts from a still portrait.
Can I make a three-minute avatar video?
The photo-to-video input limit is 180 seconds. Audio is prepared and quoted before a single generation. Longer videos may take more time; review the full result for identity and lip timing.