Skip to content

autoshow/autoshow

Repository files navigation

autoshow-bun

Overview

Full-stack application for automatically processing videos and audio with AI transcription, summarization, text-to-speech, image generation, and music generation.

Features

  • Audio Processing: Extract audio from YouTube videos, streaming platforms, or direct file uploads
  • AI Transcription: Groq Whisper, DeepInfra, YouTube Captions, or HappyScribe with automatic speaker diarization
  • LLM Summarization: OpenAI GPT, Claude, or Google Gemini models for generating episode descriptions, summaries, and chapters
  • Text-to-Speech: Convert summaries to audio with OpenAI or ElevenLabs voices
  • Image Generation: Create cover images from AI-generated prompts
  • Music Generation: Generate background music with ElevenLabs in multiple genres
  • Operational Logging: Structured JSON logs in production with request IDs, severity levels, and optional HTTP log drain forwarding
  • Docker Support: Containerized deployment with image reporting tools

Quick Start

Test URLs

https://www.youtube.com/watch?v=nXtaETBZ29g
https://ajc.pics/audio/fsjam-short.mp3
https://ajc.pics/autoshow/textract-3.pdf

Install Dependencies

bun install

Run Type Check

bun check

Start Docker Container

bun as docker up

Visit the URL printed by the CLI to generate show notes through the web interface. The default is http://localhost:4321. If port 4321 is already in use, bun as docker up prints the fallback URL it selected, or you can pin one with AUTOSHOW_DOCKER_HOST_PORT=4322.

Agent Discovery

  • /llms.txt - plain-text overview of public AutoShow pages
  • /sitemap.md - markdown index of public AutoShow pages

Documentation

CLI Commands

API Documentation

  • Overview - API endpoints and usage
  • Health - Health check endpoint
  • Jobs - Job management
  • Models - Available AI models
  • Process - Multi-step processing pipeline
  • Download - File upload endpoints
  • Media - Audio, image, and video processing

Testing

Architecture

About

End-to-end workflow to automatically generate show notes from audio/video transcripts

Topics

Resources

Stars

Watchers

Forks

Contributors