Show HN: Voice driven murder mystery, Interview AI suspects with your voice

📝 Discussion Summary (Click to expand)

Account & Access Friction – Many users say they won’t create new HN accounts or will abandon the service if they get locked out.

"sounds neat, would love to try it but I don't make accounts for websites anymore." – lorenzo95
AI Voice & API Security Concerns – Complaints about AI‑generated narration, token‑usage limits, and exposed bearer tokens in client‑side JavaScript.

"js error ... 429 (Too Many Requests) ... auth bearer token in clear js request" – zerratar
"It was working great... But now it says it can't connect. Did the server go down or did I hit one of your limits?" – londons_explore
Excitement for Voice‑Driven Interactive Storytelling – Praise for the concept’s potential, despite technical hurdles, and anticipation of its use in adventure games.

"Even without trying it, its so cool this exists." – donpdonp
"It was one of the first things I tried out with an LLM ... excited to give it a try!" – grahamburger

🚀 Project Ideas

Generating project ideas…

Eliminates the need for HN username/password accounts by letting users prove identity with a short voice clip, solving the “no‑account” frustration.
Core value proposition: instant, password‑less login/recovery for HN commenters who avoid account creation.

Key	Value
Target Audience	Hacker News commenters who avoid creating accounts and fear lockouts
Core Feature	Voiceprint authentication API that validates identity against a stored voice sample
Tech Stack	WebRTC client recording, Python backend, TensorFlow Lite model, Firebase Auth
Difficulty	Medium
Monetization	Revenue-ready: subscription per month

Directly addresses HN user complaints like “I don’t make accounts for websites anymore” and “if you got locked out of hn, you would just never comment again?”
Sparks discussion around privacy‑preserving identity verification and could become a building block for decentralized web login.

Provides mystery stories narrated by professional voice actors instead of synthetic AI voices, addressing the aversion to AI narration.
Core value proposition: premium, human‑voiced interactive fiction with dynamic text and token‑controlled usage.

Key	Value
Target Audience	Fans of interactive text adventures and podcast listeners who reject AI voices
Core Feature	Story generation combined with selectable human voice talent and a token budgeting dashboard
Tech Stack	Node.js front‑end, AWS Polly or custom high‑fidelity TTS, GPT‑4 for story drafting, DynamoDB for token accounting
Difficulty	High
Monetization	Revenue-ready: pay‑per‑minute listening or tiered subscription

Directly answers the comment “I just can't stand AI voices... I'd rather listen to bad/amateur human narration than AI voices.”
Solves the security concern about exposing OpenAI tokens by handling all API calls server‑side, eliminating clear‑text token leaks.

Enables creators to build voice‑controlled interactive mysteries with clear networking permission prompts, fixing the “won’t get past Dispatch confirmation” issue.
Core value proposition: one‑click deployment of voice‑driven RPGs with secure OpenAI realtime API proxy and token limits.

Key	Value
Target Audience	Indie game developers and tabletop RPG designers who want voice input but lack networking setup
Core Feature	Template engine that generates a web UI with voice input, handles OpenAI Realtime calls via serverless proxy, and enforces per‑user token caps
Tech Stack	SvelteKit front‑end, Cloudflare Workers backend, OpenAI Realtime API wrapper, Stripe for token purchases
Difficulty	Medium
Monetization	Hobby

Resolves the exact error reported (“POST https://api.openai.com/v1/realtime/calls 429 Too Many Requests… token in clear js request”) by providing a ready‑made proxy and permission UI.
Generates immediate discussion potential among “vibecoding” enthusiasts and RPG communities looking for voice‑first game tools.