model: ai-talking-avatar. Submit an asynchronous GoEnhance API task, store the returned img_uuid, then poll or use callbacks for completion.AI Talking Avatar API for Lip-Synced Video
Turn one portrait and one speech track into a lip-synced talking video with
Get API Key
Connect portrait and speech inputs in one request
Start with a clear portrait
Provide a public image_url up to 10MB. The speaker's face should be clearly visible, unobstructed, and roughly front-facing for a reliable source image.
Drive speech with recorded audio
Provide a public audio_url from 3 to 60 seconds and up to 50MB. Its measured duration sets both the output length and the billable duration.
Validate the exact model name, media URLs, file sizes, audio duration, and resolution before submission.
AI Talking Avatar API capabilities and limits
Exact model identifier
Set <code>model</code> to <code>ai-talking-avatar</code>; this required value selects the talking-avatar workflow.
Required portrait URL
Pass <code>image_url</code> as a public portrait URI no larger than 10MB with a visible, unobstructed face.
Required speech URL
Pass <code>audio_url</code> as public speech audio lasting 3-60 seconds and no larger than 50MB.
Optional performance prompt
Use <code>prompt</code> to describe the desired speaking performance; the documented default is <code>person talking</code>.
Two resolution options
Choose <code>540p</code> or <code>720p</code>. The documented default is <code>540p</code>.
Callback or polling
Provide an HTTPS callback or persist <code>img_uuid</code> and poll the job-detail endpoint for completion.
Pricing follows measured audio duration. A 30-second clip costs 15 tokens at 540p or 30 tokens at 720p.
AI Talking Avatar API pricing
540p pricing
0.5 token per audio second, equal to $0.01 per second at the documented $0.02 token value.
720p pricing
1 token per audio second, equal to $0.02 per second at the documented $0.02 token value.
Measured-audio billing
GoEnhance measures the transferred audio before deducting tokens; cost is measured seconds multiplied by the selected rate.
Integrate the GoEnhance API in six steps
1
1. Create an API key
Create a GoEnhance API key and send it with your authenticated API request.
2
2. Validate source media
Check portrait visibility and size, then confirm the speech file is public, 3-60 seconds, and no larger than 50MB.
3
3. Build the request
Set
model: ai-talking-avatar, add image_url and audio_url, then choose an optional prompt and resolution. 4
4. Submit the task
POST the JSON body to
https://api.goenhance.ai/api/v1/videos/generations. Do not send a duration field. 5
5. Persist and track
Save the returned
img_uuid, then poll GET /api/v1/jobs/detail or receive callback events. 6
6. Handle terminal states
Process success and failure idempotently, validate the returned video, and keep retries separate from duplicate submissions.
Build consent-aware presenter, training, localization, and creator workflows around authorized portrait and audio inputs.
Developer use cases for talking avatar video
Localized presenter videos
Pair approved presenter portraits with localized speech tracks for product explainers and regional campaigns.
Training content pipelines
Generate talking-head segments from authorized portraits and prepared narration for internal learning materials.
Personalized onboarding
Create consent-based welcome clips for product flows without building a custom lip-sync rendering stack.
Creator narration tools
Add portrait-plus-audio generation to creator products while tracking every request asynchronously.
Campaign video variants
Produce approved speech and resolution variants for review, localization, and downstream video editing.
Automated content queues
Connect task IDs and callbacks to batch orchestration without blocking the application request thread.
Answers to common request, pricing, media-limit, task-status, and integration questions.