Speech-to-Video
AI-generatedSummary
Generate a video from an audio file and an image using the Kie.ai Speech-to-Video model, optionally guided by a prompt.
Inputs
- Audio URL (required) — URL of the audio file to use for video generation.
- Image URL (required) — URL of the image to combine with the audio for the video.
- Prompt — Optional textual prompt to guide the video generation (required for Speech-to-Video).
- Reply URL — Webhook URL that will be called when the video generation task completes.
- Reply Ref — Custom reference string to pass back in the webhook callback.
- Captcha Token — reCAPTCHA token, if the API requires it.
- Wait for Completion — Boolean flag to wait for task completion before returning (polls every 3s, with a 5-minute timeout).
Output shape
A JSON array with one object per input item. Each object contains the API response with task details and result URLs after completion if waited for.
If 'Wait for Completion' is true, the node polls the API until the task completes or fails and returns the detailed result; otherwise, it returns immediately with task creation response. The output includes URLs to the generated video.
Examples
Example 1: Generate a video from speech
Provide audio and image URLs, optionally a prompt, and enable 'Wait for Completion' to receive the final video URLs in output.