James Washington
writing about archive utils free game ideas replays
Document

audio-studio design

A plan for an engine that builds kid-friendly audio shows and delivers them to a music player.

built Aug 2026 Voice in, voice out

how it works

Manifest-driven render and deliver pipeline; design only.

Document: A reference or runbook.

The code for this one lives in a private repo. This page explains the idea so you can build your own.

designtts

related

  • Artifact skill: self-describing HTML documents with embedded DSL AI skill Makes documents (roadmaps, logs, notes) as single web pages that carry their own editing instructions, so any person or AI can update them years later.
  • Hand-written script to TTS podcast episode injector Command-line tool Turns a written script into a spoken podcast episode on a private feed.
  • dsh-voice: voice in, spoken answers out for a terminal agent Plugin Talk to a terminal AI agent and have it talk back.
  • Dream Synthesis Engine Automation Reads my notes and finds surprising links between unrelated ideas, then writes new idea notes back.
  • TTS summary/completion output styles Prompt Makes the AI announce finished work out loud.
← everything I've built
© 2026 James Washington
github twitter rss