AudioBloomAI Roadmap

This is the public product roadmap for AudioBloomAI. Source code, internal release engineering notes, signing details, and model-development experiments are kept private.

Current Release

The current public release is AudioBloom v3.2.0.

v3.2.0 promotes the stabilized Milkdrop shell, final source selector, projector/Syphon workflow, Deck B crossfader, setlists, media layer controls, and local experimental audio-semantics features into one public release.

Highlights:

  • microphone/interface, all-app audio, and single-app audio capture
  • explicit source-type and target pickers
  • Milkdrop/projectM-first visual renderer
  • Deck B A/B staging, per-deck gain, blend modes, and crossfader controls
  • optional camera media layer
  • favorites, ratings, history, queued-next preset, auto-advance, and saved setlists
  • projector, clean output, and Syphon output workflows
  • setup/help copy for Microphone, System Audio Recording, and Camera permissions
  • experimental preset assistant for intent/mood-driven setlist drafts

Near-Term Focus

  • App Store launch and TestFlight feedback.
  • Workflow polish for first-run setup, permission recovery, and source selection.
  • More preset-library ergonomics for large Milkdrop collections.
  • More validation with VJ hosts such as OBS, Resolume, TouchDesigner, and VDMX.
  • Continued tuning of Deck B, projector, Syphon, and camera-layer performance.

Experimental AI / ML Direction

AI and ML work should improve the visualizer workflow instead of distracting from it. The public direction is:

  • keep audio analysis local by default
  • avoid sending microphone, system audio, camera frames, presets, or setlists to the developer
  • use semantic audio cues only where they improve transitions, preset steering, and live performance control
  • keep experimental controls clearly labeled until they are ready for default use

Planning Principles

Stabilize the flagship path first

New work should prefer:

  • preset workflow quality
  • output workflow quality
  • transition stability
  • performance and recoverability

over broad renderer experimentation.

Keep AI behind product value

AI / ML work should only move forward when it improves:

  • preset selection
  • transition quality
  • live performance control
  • library organization
  • camera-driven visuals

Keep public and private surfaces separate

The public GitHub repository is for product information, support, privacy, Pages, and wiki content. The source code and internal documents remain local and private.