AI Sound

AI-native audio editor built as a modern replacement for Audacity, with LLM integration at its core. Features multi-track editing, AI transcription, speaker diarization, semantic search, and a full MCP server for external AI assistant integration.

Solo DeveloperNew Paradigm Web DevelopmentMarch 2026 - Present4 months

View Project

About This Project

AI Sound is designed from the ground up with AI integration — not bolted on as an afterthought. It exposes all editing capabilities through both a conversational AI assistant and an MCP server, enabling AI tools like Claude Desktop to edit, analyze, and export audio projects. Supports non-destructive multi-track editing, audio effects (normalize, compress, EQ, reverb, noise reduction), word-level transcription, speaker identification, and semantic content search.

Key Achievements

Architected an MCP server that exposes all audio editing operations as composable tools — AI assistants can import, trim, mix, transcribe, and export without understanding audio internals
Built AI-powered transcription pipeline with word-level timestamps and speaker diarization that integrates directly with the editing timeline
Implemented semantic search across transcribed audio content — find segments by meaning, not just keywords
Designed a non-destructive editing engine with full undo/redo history, markers, and automation envelopes
Audio effects chain (normalize, compress, EQ, reverb, noise reduction, fade, pitch shift, speed) all accessible via both UI and MCP tool calls
Cross-project track import enabling composable audio workflows between multiple projects
LLM-agnostic architecture — runs fully local with Ollama or connects to any OpenAI-compatible cloud provider

Related Projects

AI Charts

2026

AI-powered flowchart, ERD, and swimlane diagram builder with a built-in AI assistant and an MCP server exposing 18+ tools for external AI integration.

TypeScriptReactNode.js +4

eRepublic Registration Management System (ERMS)

2017

ERMS is a Windows and OSX desktop application that allows the management of every aspect of the eRepublic event registration process.

ElectronTypeScriptJavaScript +4

RSVP

2024

RSVP is a new approach to recorded interviews allowing a group of editors to create an interview series for users to complete and share with their audience.

ReactTypeScriptCapacitor +3

View all projects ->

Related Publications

Articles that cover technologies and skills used in this project

Prop Drilling is a Code Smell When Used Incorrectly

[M]medium

Oct 10, 20244 min read

Prop Drilling is a Code Smell When Used Incorrectly

Prop drilling is considered an anti-pattern in React because it involves passing props down multiple levels of a component tree, even when only a few components need them. This can lead to several...

RPC; Everything Old is New Again

Possibly since the inception of computing, the paradigm of client to server communication has been an important, sometimes overlooked requirement for distributed computing. From a developer...

Benchmarking Models in the Real World

I tested a bunch of SOTA LLM models on real world agentic tasks that go beyond coding and typical business processes and really stress them. There were some shocking...