Yes, you can make AI videos with sound on Wurt.app. In Vidwurt, choose a model with sound, such as Sqwurt (New), Horse or Seedance 2.0, and describe the voices, music and effects in your prompt. The sound is made in the same generation as the picture, so footsteps, speech and music land with the action. Not every option makes audio: GIFs, for example, are silent.
Make a video with soundOn models with sound, the audio is created together with the video, not pasted on afterwards.
Write the exact line in quotes and say who speaks it and how.
Name the style, the mood and when it starts, or ask for no music at all.
Rain, engines, crowds, a door slam: describe the sound and the moment it happens.
Seedance 2.0 and Edit Any Video accept audio clips to guide a voice, a beat or a sound.
Ask for "no music", "no talking" or a silent clip, and the prompt carries that direction.
Sound depends on the model you pick. Here is what each option does today:
| Option | Sound | Good for |
|---|---|---|
| Sqwurt (New) | Makes sound with the video | Mature content and the longest clips, up to 30 seconds |
| Horse (under Professional) | Makes sound with the video | Mature scenes with several reference images |
| Seedance 2.0 (under Professional) | Makes sound with the video; accepts reference audio | Reference-heavy and multi-shot work |
| Wurt | Adds sound when your prompt asks for it | Everyday text-to-video and image-to-video |
| Quick Animate and Spicy Animate | Clips come with sound | One-tap animation of a Picwurt picture |
| Edit Any Video | The edited clip comes with sound; accepts up to 3 audio clips | Changing a video you already have |
| Create Any GIF | No sound (it makes GIFs) | Short looping animations |
The rule is simple: if sound matters, pick a model from the top of this table and describe the sound you want.
The model can only follow sound you describe. Be concrete, and tie each sound to a moment in the clip.
Say what you do not want, too: "no music", "no talking" or "silent" are all valid sound directions. More prompt tips are in how to prompt AI video.
Two tools take audio files as references. Seedance 2.0 has a reference audio slot: attach a clip and point to it in your prompt with @Audio1 (then @Audio2 and @Audio3 for more). Edit Any Video lets you add up to 3 audio clips next to the video you want to change. Use the prompt to say how each clip should be used, for example as the music bed or as a voice to match.
Many older AI video tools made silent clips, so you had to add music or effects in a separate app. In 2026 the leading video models, including Veo 3.1, Kling 3.0 and Seedance 2.0, create audio in the same pass as the picture. That keeps lips, footsteps and impacts in sync with the motion.
Adding sound later still makes sense for long edits with a licensed soundtrack or a voice-over. For short clips, a model with sound saves a whole step.
Want more length for dialogue? Sqwurt (New) makes single clips up to 30 seconds. See the long AI video generator guide, or start from a picture with animate a photo.
Yes. Choose a Vidwurt model with sound, such as Sqwurt (New), Horse or Seedance 2.0, and describe the dialogue, music and effects in your prompt. Wurt adds sound when your prompt asks for it.
Create Any GIF makes GIFs, and GIFs are silent. If you need audio, pick a video model with sound instead.
Yes, on models with sound. Put the exact line in quotes and say who speaks it and how, for example warmly or in a whisper.
Seedance 2.0 accepts reference audio clips, which you name in the prompt as @Audio1, @Audio2 and so on. Edit Any Video accepts up to 3 audio clips next to your video.
Ask for it in the prompt with "no sound" or "silent", or name the sounds to leave out, such as no music or no talking. For a clip with no audio track at all, Create Any GIF makes silent GIFs.
Pick a model with sound, describe the audio, and create. Start with free starting credits.
Make a video with sound