Polish dialogue and lip sync with Kling 3.0

Evaluate lip sync using both picture and sound. Convincing mouth motion is not enough if the sentence is hard to understand. Start with a fictional speaker and one short line. This workflow generates dialogue with video; it does not replace a real person’s voice.

Prepare the shot

Write the exact Polish line in quotation marks. Start without proper names or numbers.
Use one fictional speaker facing the camera. Keep hands and objects away from the mouth.
Enable generated audio and check the updated estimate before submitting.
Request Polish speech, a calm pace and no music.
Listen on headphones, then watch silently. Inspect the beginning and end of the utterance.
Add manually checked captions and verify the meaning of any advertising claim.

Review before publication

Every word is intelligible.
Mouth motion begins and ends with speech.
Captions match the audio.

API input example

The JSON below is the request body for POST /v1/generate. Use your own accessible image URL where a reference is required.

json

{
  "model": "kling-3.0",
  "input": {
    "prompt": "A fictional animated shopkeeper facing the camera says in Polish: \"Dzień dobry, zapraszam do środka.\" Calm speech, clear mouth movement, no music, no subtitles",
    "duration": 5,
    "generate_audio": true
  }
}

Model and cost

Use Kling 3.0. Its current unit price is from $0.169 per second. Check the pricing page and calculator for your settings. Credits never expire, there is no subscription and failed generations are free. A successful output is billable even if you choose not to publish it.

FAQ

Does enabling audio guarantee correct Polish dialogue?

No. It enables audio generation, but pronunciation and synchronization still require review.

Can I check the request before a paid run?

Yes. Check the model schema and the cost shown in Studio. Local demo API tests validate request handling but do not prove the visual or audio quality of the real model.

Where are the supported parameters?

Model details and the generation API guide list the supported inputs. Preview mode does not provide a real generated result.