Prompt library

Dialogue & lip sync

Street interview, vertical

TEXT → VIDEO8s9:161080P · AUDIO ON
Vertical street-interview shot: a young man holding nothing, framed waist-up on a busy daytime shopping street, looks slightly off-camera and says in Japanese: "正直、思ったより全然よかったです。" He laughs at the end. Handheld energy, passersby blurred behind him. Audio: his voice, street crowd, light wind on mic.
Why it works

The fake vox-pop is the highest-performing ad format on vertical feeds, and it lives or dies on imperfection: off-camera eyeline, a laugh, wind on the mic. The Japanese script exercises HappyHorse's multilingual lip sync where competitors fall back to dubbing.

The breakdown
Format
9:16 waist-up vox-pop — native feed grammar
Eyeline
slightly off-camera — interviews look at the interviewer, not the lens
Script
Japanese line in quotes + "laughs at the end" — speech plus a non-verbal beat
Audio
voice + crowd + mic wind — deliberately imperfect mix
Field notes
  • The laugh is the money frame — non-verbal beats after the line make the sync read as real.
  • Ask for "handheld energy", not "shaky camera"; the first is style, the second is a defect.
Related prompts