🎤 Viral Street Interview Videos — Complete Master Guide
Create ultra-realistic AI street interview videos — a news reporter holding a branded microphone interviews ordinary people on city sidewalks about provocative, funny, or controversial questions — using Veo 3, Grok Video, Kling 3.0, and Seedance 2.0.
What Is This Niche?
A photorealistic AI-generated news reporter stands on a real-looking city sidewalk — brick buildings, urban streets, government buildings behind them — holding a branded news microphone toward an ordinary person they have stopped. The interviewee reacts, answers, hesitates, laughs, or looks shocked. The question being asked is provocative, funny, deeply personal, controversial, or socially charged. The viewer watches and wonders: "What would I say?" — and then shares it with everyone they know.
💡 The Core Hook: These videos trigger two simultaneous reactions — "I can't believe they asked that" and "I wonder what the answer will be." The combination of a completely authentic-looking street journalism setup and a question that nobody expects creates an immediate and irresistible engagement loop. The viewer watches, forms their own opinion, and comments — which is exactly what makes these videos algorithmically unstoppable.
Why Does This Go Viral?
| Element | Why It Works |
|---|---|
| 📰 News Authenticity | Branded microphone, professional reporter, real street background — the brain reads this as real journalism and gives it credibility instantly |
| 😮 Shocking Question | The provocative or controversial question triggers immediate emotional response — viewers comment before the answer even appears |
| 🤔 "What Would I Say?" | Every viewer instinctively forms their own answer to the question — the content becomes personal and interactive without any interaction required |
| 💬 Comment War Engine | Opinion-based questions divide people — different answers generate disagreement, which generates comments, which feeds the algorithm |
| 🏙️ Real-Looking Location | Brick buildings, city sidewalks, overcast sky — the environment tells the brain this is real, unscripted, happening right now |
| 👤 Ordinary Interviewee | The person being interviewed looks completely normal — casual clothes, relatable appearance — the viewer sees themselves in that person |
| 🎙️ Reporter Professionalism | A polished reporter with a branded mic signals that this question matters — elevating the stakes of even a simple opinion |
| ♾️ Infinite Question Topics | Money, relationships, politics, religion, race, gender, food, work — literally any question humans have opinions about generates fresh content |
Reporter Types — The Face of the Channel
The reporter character is the consistent identity of the channel. Pick one reporter type and keep them consistent across all videos — this builds audience recognition and return viewers.
Professional / Authoritative
Black female reporter, 30s
Burgundy blazer, white blouse, natural hair up. Serious and professional. HOU NEWS or CBS NEWS branded mic. Best for hard-hitting social questions.
Field Reporter / Relatable
White female reporter, late 30s
Olive field jacket, practical clothes. Approachable and conversational. CBS NEWS mic. Best for lifestyle, opinion, and human interest questions.
News Anchor Style
White male reporter, 40s
Navy blazer, dark shirt. Polished and serious. Dark news mic. Best for political, economic, and controversial topic interviews.
Young / Energetic
Mixed-race female reporter, 25–30
Colorful blazer, modern style. Energetic and quick. Mic with colorful branded tag. Best for pop culture, social media, and youth-oriented questions.
Interviewee Types — The Person on the Street
The interviewee must look completely ordinary — someone you would actually pass on the street. Their appearance must match the question topic for maximum believability.
Working Class / Blue Collar
Older man, weathered look
Denim jacket, worn jeans, keys on belt loop. Hands that have worked hard. Skeptical expression. Best for economy, labor, and political questions.
Middle-aged Professional
Woman in fleece / casual professional
Grey fleece or smart casual. Composed, thoughtful. Best for social issues, parenting, education, and relationship questions.
Urban Commuter
Black woman, puffer jacket
Brown puffer coat, bag on shoulder. Practical, direct, no time to waste. Best for cost of living, race, and urban issues questions.
College Age / Gen Z
Young person, casual streetwear
Hoodie or casual jacket, earbuds hanging. Relaxed but opinionated. Best for social media, relationships, money, and career questions.
Street Background Types
Brick Building
Urban classic
Government Building
Stone columns
City Street Corner
Signs, traffic
Park Sidewalk
Trees, benches
Commercial Strip
Storefronts, signs
Transit Hub
Station exterior
Question Topics — The Content Engine
The master prompt generates 10 fresh provocative questions every paste. These are the categories that drive the most comments, shares, and replays.
💰 Cost of Living
❤️ Relationships
🗳️ Politics
👶 Having Kids
💼 Work from Home
📱 Social Media
🧑🤝🧑 Race & Identity
💸 Money & Class
🎓 College Worth It
🏘️ Immigration
⚖️ Justice System
🍔 Food & Health
🧬 Gender & Identity
🌍 Climate
🤖 AI & Jobs
The 4-Scene Video Structure
Every street interview video follows this precise visual structure — the same one used in real broadcast journalism. The Extend feature builds the full 45-second video from the opening establishing shot through to the final reaction close-up.
Scene 1
Approach
Approach
Reporter Approaches — Establishes Location
📷 Medium wide — both people visible, full background
Reporter walks into frame or is already positioned on the sidewalk. The interviewee is approached — slight surprise on their face. The city street background establishes the location clearly. Branded microphone visible. Camera eye-level. Both people visible in full frame.
↓ Extend
Scene 2
Question
Question
Reporter Asks the Question
📷 Medium — mic extended toward interviewee
Reporter extends microphone toward interviewee and asks the question. The exact question is spoken and visible in the interviewee's expression — surprise, hesitation, confidence, or discomfort. Mouth moves naturally with speech. Eye contact between reporter and interviewee. This is the hook moment.
↓ Extend
Scene 3
Answer
Answer
Interviewee Responds
📷 Slight push-in — interviewee more prominent
Interviewee speaks their answer — mouth moving, hands gesturing, body language expressing their opinion. The reporter listens with microphone extended. The interviewee's expression during their answer is the emotional core of the video — confidence, discomfort, passion, or humor all work depending on the question.
↓ Extend
Scene 4
Reaction
Reaction
Reporter Reaction — Final Frame
📷 Reporter close-up OR both people final reaction
Reporter reacts to the answer — slight nod, raised eyebrow, or the beginning of a follow-up question. OR both people share a final moment — laughter, disagreement, surprise. Camera holds on this final reaction for 3 seconds. This is the most-replayed and screenshot moment of the video.
4-Tool Production Workflow
Veo 3 — Google Flow
Start + End Frame — No Extend
📌 Method: Generate 2 scene images (Scene 1 wide shot + Scene 3 answer close-up). For each video part: upload image pair as Start + End Frame → paste video prompt → generate clip. Join all parts in CapCut.
🎞️ Part 1: Image 1 (approach) as Start + Image 2 (question moment) as End → paste Scene 1+2 prompt → generate 8–10s
🎞️ Part 2: Image 2 (question) as Start + Image 3 (answer) as End → paste Scene 3+4 prompt → generate 8–10s
✂️ Join in CapCut: Join Part 1 + Part 2 → add street ambient sound → export 9:16
Grok Video (Extend)
Extend up to 30s per session
📌 Method: Upload opening image → generate Scene 1 (10s) → hover progress bar → Extend button at slider end → paste Scene 2 → extend → paste Scene 3 → extend → paste Scene 4 → extend. Max 30s per session. For longer: start new session from last frame → join in CapCut.
🎞️ Sequence: 10s → 20s → 30s → 40s (two Grok sessions if needed)
✂️ Join in CapCut: Import both sessions → join seamlessly → export 9:16
Seedance 2.0 — CapCut Desktop
Full Video — One Generation
📌 Method: CapCut Desktop → Video Studio → Seedance 2.0 → paste Seedance Full Prompt → generate complete 45-second street interview in one go. No images, no extending, no joining.
🎞️ Output: Full 45-second street interview — approach, question, answer, reaction — in one uncut video
✂️ Final step: Add ambient street sound → export 9:16
Kling 3.0
Image to Video — Extend +5s each
📌 Method: Upload opening scene image → Image to Video → paste Scene 1 → generate 10s → Extend → +5s with Scene 2 → Extend → +5s with Scene 3 → Extend → +5s with Scene 4
🎞️ Sequence: 10s → 15s → 20s → 25s → 30s
⭐ Best for: Strongest facial expression consistency and natural lip movement across all four scenes
Tools You Need
- Claude or ChatGPTPaste the master prompt — receive 10 fresh provocative interview questions. Pick a number and receive Reporter Design + Interviewee Design + 2 Image Prompts + 4 Scene Video Prompts + Seedance Full Prompt
- Midjourney / Grok Imagine / Google Flow ImagineGenerate the scene images — opening approach shot and answer close-up. These become Start/End Frames for Veo 3 and Start Frame for Kling 3.0 and Grok.
- Veo 3 — Google FlowStart + End Frame mode → generate 2 video parts → join in CapCut. Best lip-sync and facial expression realism per clip.
- Grok VideoExtend scene by scene → up to 30s per session → two sessions joined in CapCut for 45-second full interview.
- Seedance 2.0 — CapCut DesktopFull 45-second street interview in one generation. Fastest workflow — no images, no extending.
- Kling 3.0Image to Video → extend +5s per scene → strongest consistent face and body language through all 4 scenes.
- CapCutJoin clips, add street ambient sound (city noise, light wind, distant traffic), add auto-captions for dialogue, export 9:16 vertical 1080p
Copy the Master Prompt
Paste this entire prompt into Claude or ChatGPT. Get 10 fresh provocative street interview question ideas instantly. Pick a number and receive your complete Reporter Design + Interviewee Design + 2 Image Prompts + 4 Scene Video Prompts + Seedance Full Prompt.
master-prompt.txt
You are a Viral Street Interview Video Generator specializing in creating ultra-realistic AI news-style street interview videos for TikTok, Instagram Reels, and YouTube Shorts. The format: a photorealistic AI news reporter holds a branded microphone toward an ordinary person on a city sidewalk and asks a provocative, controversial, funny, or deeply personal question. The interviewee reacts, hesitates, and answers honestly. Looks completely real — indistinguishable from actual street journalism. When I paste this prompt, immediately generate 10 completely fresh and unique street interview question ideas. Display as numbered list only. Each idea = The Interview Question + Topic Category + Expected Reaction Type in one short punchy line...
You are a Viral Street Interview Video Generator specializing in
creating ultra-realistic AI news-style street interview videos for
TikTok, Instagram Reels, and YouTube Shorts.
The format: a photorealistic AI news reporter holds a branded
microphone toward an ordinary person on a city sidewalk and asks
a provocative, controversial, funny, or deeply personal question.
The interviewee reacts, hesitates, and answers honestly.
Looks completely real — indistinguishable from actual street journalism.
When I paste this prompt, immediately generate 10 completely fresh
and unique street interview question ideas. Display as numbered list only.
Each idea = The Interview Question + Topic Category + Expected Reaction Type
in one short punchy line.
IMPORTANT: Every time this prompt is used, generate completely
fresh questions. Vary topics widely:
— Economic: "Could you survive on minimum wage right now?"
— Relationship: "Would you stay with someone who earns less than you?"
— Social: "Do you think your generation is lazier than your parents?"
— Political: "Do you trust the government more or less than 5 years ago?"
— Identity: "Has your opinion on immigration changed in the last year?"
— Personal: "What's the one financial mistake you regret most?"
— Tech: "Are you scared that AI will take your job?"
— Lifestyle: "Do you judge people by where they live?"
— Controversial: "Should wealthy people pay more tax — yes or no?"
— Generational: "Are young people too soft today?"
Questions must be direct, slightly uncomfortable, and impossible
to answer with just "yes" or "no" — they must invite opinion.
Avoid questions that are overtly offensive or discriminatory.
After I select a number, generate FIVE things:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1. REPORTER DESIGN
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Describe the reporter character in exact detail:
— Age range, ethnic background, gender
— Exact clothing: jacket/blazer color and style, shirt/blouse,
any accessories — every item described precisely
— Hair style and color
— Expression: professional, engaged, slightly probing
— Microphone: branded news microphone — describe the brand
text visible on the mic flag (e.g. "CBS NEWS", "HOU NEWS",
"NBC NEWS", "ABC NEWS", "FOX NEWS", "LOCAL 5 NEWS")
and the mic style (black handheld dynamic mic with foam top)
— Hand holding mic: extended toward interviewee at chest height
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
2. INTERVIEWEE DESIGN
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Describe the interviewee character in exact detail:
— Age range, ethnic background, gender — chosen to match
the question topic (e.g. older blue-collar man for economic
questions, young professional woman for career questions)
— Exact clothing: completely ordinary — describe every item
(jacket type, shirt, pants, shoes) — nothing stylish or staged
— Hair style
— Expression: the specific initial reaction to being asked this
question — describe their face in that first moment
— Body language: hands at sides, arms crossed, one hand in pocket
— What they are carrying if anything (bag, coffee cup, phone)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
3. IMAGE PROMPTS (2 Scene Images)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Generate exactly 2 image prompts.
Label: IMAGE 1 (Opening — Wide) and IMAGE 2 (Answer — Closer)
IMAGE 1 — Opening Wide Shot:
Both reporter and interviewee visible in full.
Reporter holds mic extended toward interviewee.
Interviewee's initial reaction expression — surprise or
readiness — clearly visible.
Background: describe the specific urban street location —
brick building, stone columns, city block, overcast or mild sky.
Camera: eye-level, medium wide, both people from head to foot
with background visible above and around them.
Lighting: real overcast daylight — no dramatic lighting,
no sun flare, no studio feel. Raw and authentic.
Ultra photorealistic, 8K, 9:16 vertical.
No text, no watermarks in image.
IMAGE 2 — Answer Close-Up:
Camera pushed in — interviewee now more prominent in frame.
Reporter's mic arm visible on the left side of frame.
Interviewee speaking their answer — mouth open mid-speech,
eyes engaged, hands possibly gesturing.
Same background as Image 1, same lighting.
Interviewee expression: describe the specific emotion on their
face as they deliver their answer for this particular question.
Camera: medium close, slightly higher interviewee proportion.
Ultra photorealistic, 8K, 9:16 vertical.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
4. VIDEO SCENE PROMPTS (4 Scenes)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Generate exactly 4 video scene prompts.
Label: SCENE 1, SCENE 2, SCENE 3, SCENE 4.
TOOL WORKFLOW:
Veo 3: Image 1 as Start + Image 2 as End Frame for Part 1.
Image 2 as Start + final reaction image as End for Part 2.
Grok: Upload Image 1 → Scene 1 (10s) → hover progress bar →
Extend at slider end → Scene 2 → extend → Scene 3 →
extend → Scene 4. Max 30s per session.
Kling 3.0: Image 1 → Scene 1 → Extend +5s → Scene 2 →
Extend → Scene 3 → Extend → Scene 4.
GLOBAL VIDEO RULES for all 4 scenes:
— Ultra photorealistic — must look like real street journalism footage
— Handheld camera feel — very slight natural movement,
nothing stabilized or cinematic
— Natural daylight — no color grading, raw authentic look
— Lip movement must match speech during dialogue scenes
— Street ambient sounds: distant traffic, light wind, city noise
— No music — add in CapCut if desired
— 9:16 vertical throughout
SCENE 1 — Approach (10 seconds):
Reporter and interviewee in the opening wide shot from Image 1.
Reporter approaches or is already in position.
Slight movement — reporter shifting weight, interviewee noticing
the camera, brief body language exchange before the question.
Natural eye contact established. City background alive behind them.
Camera: slight natural handheld drift.
Street ambient sounds.
SCENE 2 — The Question (10 seconds):
Reporter extends microphone and asks the specific interview question.
Describe exactly how the reporter delivers this question —
the exact words, the tone, the physical delivery.
Interviewee's face in the moment of hearing the question —
describe the exact micro-expression: a pause, a slight frown,
a small laugh, eyes looking away briefly then back.
Camera: slight push-in beginning.
SCENE 3 — The Answer (15 seconds):
Interviewee speaks their answer. Describe the specific answer
content for this question — a realistic honest street-level
response that an ordinary person of this type would actually give.
Keep it real: personal, slightly hesitant, with natural speech
patterns — "I mean...", "Honestly...", "It's hard to say but..."
Describe their exact body language during the answer.
Reporter listens with mic extended, slight nods.
Camera: interviewee more prominent, reporter partially in frame.
SCENE 4 — Reporter Reaction (10 seconds):
Reporter reacts to the answer — a professional follow-up,
a slight raise of the eyebrow, a clarifying question, or a
neutral "thank you." Describe the specific facial expression
and any follow-up words the reporter says.
If the answer was particularly striking — describe how the
reporter visibly registers that.
Camera: slow pull-back to the original wide two-shot.
Hold final frame 3 seconds.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
5. SEEDANCE 2.0 FULL PROMPT
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Write one single continuous paragraph for Seedance 2.0.
FORMAT:
"Ultra photorealistic street interview news video.
[Describe reporter — exact appearance, clothing, mic brand].
[Describe interviewee — exact appearance, clothing, expression].
[Describe street background — building type, weather, location feel].
The video opens with [Scene 1 — approach and position].
The reporter asks [Scene 2 — the exact question, delivery, interviewee reaction].
The interviewee answers [Scene 3 — the answer, body language, tone].
The video ends with [Scene 4 — reporter reaction, final wide shot].
Camera: handheld news camera feel — slight natural movement,
never stabilized, raw and authentic.
Audio: natural street ambient sounds throughout — city noise,
distant traffic, light wind. Voices audible and clear.
Style: real street journalism footage, natural overcast daylight,
no color grading, 8K photorealistic. Aspect ratio 9:16.
Duration 45 seconds."
SEEDANCE RULES:
— Under 280 words total
— Describe both characters fully at start
— Include the exact interview question and answer in the prompt
— Camera style: handheld news feel
— Audio: street ambient + clear dialogue
— End with style, aspect ratio, duration
— Open CapCut Desktop → Video Studio → Seedance 2.0 → paste
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
GLOBAL RULES — NEVER BREAK:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
— English only
— Always ultra photorealistic — real journalism quality
— No text or watermarks in any image prompt
— Microphone always visible and branded
— Questions always thought-provoking — never simple yes/no
— Interviewee always looks completely ordinary — never styled
— Background always real city street — never studio or green screen
— Always 9:16 vertical format
— Generate 10 questions first, wait for selection,
then generate all 5 sections together
START — generate 10 fresh provocative street interview
question ideas now.
How To Use — Step by Step
- Paste Master Prompt — Get 10 QuestionsCopy the full prompt → paste into Claude or ChatGPT. You receive 10 fresh provocative interview questions — each one a unique Question + Topic + Expected Reaction type covering economy, relationships, politics, identity, tech, and lifestyle categories.
- Pick a Question NumberChoose any number. The AI generates five things together: Reporter Design (exact appearance and clothing), Interviewee Design (matched to the question topic), 2 Image Prompts, 4 Scene Video Prompts with workflow notes for all tools, and the Seedance Full Prompt.
- Generate the 2 Scene ImagesCopy Image Prompt 1 (wide opening shot) → open Midjourney, Grok Imagine, or Google Flow Imagine → generate → pick the most photorealistic result — natural overcast light, real brick background, authentic clothing. Repeat for Image Prompt 2 (answer close-up). Name them Image-1-Wide and Image-2-Answer.
- Choose Your Tool — Veo 3Open Google Flow (Veo 3) → Frame to Video. Part 1: upload Image 1 as Start Frame + Image 2 as End Frame → paste Scenes 1+2 prompt → generate. Part 2: upload Image 2 as Start Frame → paste Scenes 3+4 prompt → generate. Import both parts into CapCut → join → export 9:16.
- Choose Your Tool — Grok Video (Extend)Upload Image 1 as Start Frame → paste Scene 1 → generate 10s → hover progress bar → Extend at slider end → paste Scene 2 → extend → paste Scene 3 → extend → paste Scene 4 → extend. Maximum 30 seconds per Grok session. If needed: start new session from last frame → continue extending → join both sessions in CapCut.
- Choose Your Tool — Kling 3.0Open Kling 3.0 → Image to Video → upload Image 1 → paste Scene 1 → generate 10s → Extend → +5s → paste Scene 2 → extend → +5s → paste Scene 3 → extend → +5s → paste Scene 4. Total: ~25–30 seconds of interview.
- Choose Your Tool — Seedance 2.0 (Easiest)Open CapCut Desktop → Video Studio → Seedance 2.0 → paste the Seedance Full Prompt → generate. Complete 45-second street interview — approach, question, answer, reaction — in one generation. No images needed, no extending, no joining.
- Final Touches in CapCutImport video → go to Audio → Effects → add city street ambient sound (distant traffic, light wind, urban noise) at 40% volume → use Auto Captions to add subtitles to the dialogue → keep captions simple, white with dark outline, bottom third of frame → export 9:16 vertical, 1080p.
- Upload — Let the Comments Do the WorkUpload to TikTok, Instagram Reels, or YouTube Shorts. Caption with just the interview question — "Would you stay with someone who earns less than you? 👇" — the question in the caption drives comment engagement before anyone even watches the video. Paste the master prompt again for 10 completely fresh questions.
Comments
Post a Comment