0xwilliamortiz
Featured
openclaude-improved
OpenClaude is an open-source CLI coding agent that runs on any platform and supports a wide range of LLM providers, offering the same tools and workflows as Claude Code.
project-echo
Project ECHO is an open-source AI pipeline from Commonwealth Bank of Australia that transcribes audio and generates vision-model scene descriptions from archival video footage, turning collections such as VHS and Betamax tapes into a searchable knowledge base.
Stars
4
7-day growth
No data
Forks
0
Open issues
0
License
MIT
Last updated
2026-07-31
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It is a rare instance of a major bank open-sourcing an applied AI/media-archival system, combining Whisper transcription, VLM-based frame descriptions, optional LLM summarisation, and configurable local/S3 storage with SQLite/DynamoDB backends.
Who it is for
- Archivists and digital-preservation teams managing legacy magnetic-tape collections
- Developers building searchable video or audio archives
- ML engineers deploying self-hosted OpenAI-compatible Whisper, vision, and LLM endpoints
- Organisations with large video libraries needing batch cataloguing and metadata extraction
Use cases
- Transcribe and visually describe digitised VHS, Betamax, and other magnetic-tape footage
- Create timestamped transcripts and keyframe scene descriptions for archive catalogues
- Batch-process media from local filesystems or S3 into SQLite or DynamoDB
- Generate multi-modal Markdown summaries of footage when summarisation is enabled
Strengths
- Storage and database backends are independently configurable, so it can run from a laptop with sufficient video memory to production infrastructure
- Designed for any OpenAI-compatible service, including self-hosted vLLM deployments
- Includes operational safety nets such as max file size, max Whisper chunks, max frames per video, dry-run, and refresh controls
- Provides tested code with mocked external services and an enforced 80% coverage, plus a clear CLI and configuration reference
Considerations
- Requires external OpenAI-compatible transcription and vision endpoints, so it is not a fully self-contained offline pipeline
- Published for research, educational, and community purposes only; it is not a CBA production system, is not advice, and is provided 'as is' without warranty
- Default model names such as gpt-4o and gpt-5.2 and endpoint settings must be adjusted to the models actually deployed; the README does not describe an end-user search UI
README quick start
Quick Start
Related repositories
Similar projects matched by category, topics, and programming language.
deerwork-ai
Featured
deer-workflow
An open-source Dynamic Workflow runtime that combines deterministic TypeScript orchestration with replaceable Agent runtimes.
jamesob
Featured
local-llm
A comprehensive guide for building and configuring a high-end local machine to run state-of-the-art LLMs, with detailed hardware choices, BIOS tuning, and Docker-based model serving.