Direct answer
The finished MP4 includes the 15-second scene soundtrack: the elevator hum, the headphone leak, the voices, the bell and the doors. It is mixed after the take renders. The audio is fixed; the site does not clone your voice or let you upload a track.
Make my videoSound and audio
The scene's 15-second soundtrack is mixed after the render. No voice cloning, no generated lyrics, no custom dialogue.
Published November 10, 2026
What is in the file
- A low elevator room tone under the whole scene.
- A thin, leaky headphone bed — the reason the other person can hear the music.
- The spoken beats, timed to the take.
- A bell and a door slide near the end of the take.
- Room tone under the held look.
What is not in the file
No licensed Smiths recording and no lyrics. No clone of either person’s voice. No custom dialogue or uploaded music. The audio is a fixed parody bed, and the site does not claim otherwise.
Posting with the trend sound
- 1Download the MP4 from the result page or My creations.
- 2Open TikTok, Reels, Shorts or YouTube and add the video. The MP4 already has the scene audio.
- 3Pick the “I love the Smiths” trend sound from the platform’s audio page — not a re-upload.
- 4Set the sound volume so the line lands at 3–7 seconds; the take is timed for that beat.
- 5Label the post as AI-generated, and keep the parody/non-affiliation note in the caption or pinned comment.
Why the audio is added afterwards
Video models generate motion, not a reliable soundtrack. Asking one to speak a specific line produces a different voice and drift. Mixing a fixed bed after the render keeps the timing exact and the file small. It also means the same video can be re-posted with a platform sound later.