Voice driven murder mystery, Interview AI suspects with your voice
Details
- External ID
- 49238851
- Source
- HN
- Company
- —
- Product
- Voice driven murder mystery, Interview AI suspects with your voice
- Website domain
- whodunnitai.com
- Launched
- Aug. 10, 2026
- Cohort
- —
- Upvotes
- 215
- Upvotes percentile
- 0.9731182795698925
- Tags
- —
- Fetched at
- Sept. 10, 2026, 5:32 a.m.
- Updated at
- Sept. 10, 2026, 5:32 a.m.
Description
Hey HN!I'm excited to show off this really fun project I put together. I originally built this project 2-3 years ago, AI was already booming at the time, however voice AI agents were still very early. I loved my proof of concept at the time, but wasn't quite happy with it.I recently had the desire to check out the tech again, and know many of you will be interested.Interviews are speech to speech with OpenAI's gpt-realtime-2.1 over WebRTC. This model is... expensive, and because of that, I have to add some amount of restrictions, conversations are tied to a authenticated Clerk user id. I have also added a 30 minute timer because well, I really don't want to go broke while I sleep tonight.Each suspect has a tool they call when you make a direct accusation. It captures who you accused and a faithful list of the evidence you actually stated.A separate gpt-5-mini judge then decides which of the case's required evidence facts you genuinely presented. Paraphrasing counts, vague suspicion and fishing don't.The rest is Next.js, MongoDB, and Clerk.Let me know whether the suspects hold up under a real interrogation.
Enrichment
- Theme
- task-specific ai agents and assistants
- Vertical
- Media & entertainment
- Function
- Content generation
- Audience
- B2C
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- voice-interactive murder mystery game
- Manually corrected
- False
Could you build this?
Yes The game uses OpenAI's Realtime API (gpt-realtime-2) via WebRTC/WebSockets with predefined suspect system prompts and persona rules. Creating an interactive voice web frontend connected to existing voice-agent APIs is well within the capabilities of AI-assisted coding.
Discussion
20 comments analyzed.
Competitors mentioned: The Golden Blade (Diplomacy card game), Supertonic 3 (for text-to-speech)
Concerns raised: Voice isolation picks up own voice when airplaying to group, Voice replies sometimes stop mid-response despite text being present, Characters have inconsistent knowledge base (won't answer dinner questions but knows medical advice), AI portraits obviously look AI-generated, Real-time voice interruptions and pause handling not fully solved
Feature requests: Mute toggle as keyboard shortcut for group discussion, Accusation flow: button to ready → voice to accuse → AI asks why → step-by-step evidence presentation, User stats when logged in (time to complete, etc), More involved/complex mysteries, Lightweight TTS model like Supertonic 3 for web inference
Competitors
Other products that read as similar to this one — 165 launches clear the similarity bar, closest 8 shown.
Attention rank: #9 of 166 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 284 days after the earliest competitor.
- Open-source simulation testing infra for voice agents · hn · 2026-09-10 · 16 upvotes · similarity 0.46
- I built a voice AI that responds like a real woman · hn · 2026-03-25 · 6 upvotes · similarity 0.46
- Teapot · hn · 2026-02-18 · 7 upvotes · similarity 0.46
- VoxConvo · hn · 2025-11-07 · 10 upvotes · similarity 0.44
- Echologue · hn · 2026-07-29 · 29 upvotes · similarity 0.43
- Talkstack · ph · 2026-09-29 · 2 upvotes · similarity 0.42
- InterviewChamp.AI · ph · 2026-09-28 · 1 upvotes · similarity 0.41
- sample-voice-ai-agent-studio · github · 2026-09-24 · 113 upvotes · similarity 0.40
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a content generation tool for Government yet.