Advanced Topics All In One

All-in-One Integration

AIBudsSDK/AllInOne is the convenience integration for products that want the complete SDK suite. One dependency brings in Core device support, ABMate connectivity, logging, AI providers and dashboard, on-device voice assistant authentication, crash reporting, live-stream middleware, audio dependencies, and video stabilization.

Use Installation and Quick Start for the install and initialization code. This page explains the architectural consequences of that choice.

Unified suite initialization

All-in-One selects the bundled implementations, initializes Core and AI in order, then starts shared audio coordination.

1. Configure Logging — Attach the bundled XLFacility logging destination.
2. Register Auth Bridges — Install StarBurst, MLTCloud, and on-device voice assistant bridges.
3. Register Stabilization — Install the built-in optional video post-processing plugin.
4. Initialize Core — Initialize AIBudsSDK with ABMate, configuration, and delegate.
5. Initialize AI — Register the bundled StarBurst and MLTCloud provider SDKs.
6. Start Audio Coordination — Start the shared AudioSessionCenter after both initialization stages succeed.
7. Suite Ready — Allow feature owners to begin capability-specific work.

What Initialization Owns

AIBudsAllInOneSDK.initialize performs coordinated setup rather than simply forwarding to Core:

1. Configures the XLFacility logging plugin.
2. Registers StarBurst and MLTCloud device-auth bridges.
3. Registers the on-device voice assistant authentication bridge.
4. Registers the built-in video stabilization plugin.
5. Initializes AIBudsSDK with ABMateSDK and the supplied Core configuration/delegate.
6. Initializes AIBudsAISDK with the bundled StarBurst and MLTCloud provider implementations.
7. Starts the shared audio-session coordinator after successful initialization.

Crash Reporter installation and AI Dashboard startup remain explicit operations; they are included but are not automatically enabled by initialize.

When All-in-One Fits

• The product expects to expose most SDK feature families.
• Fast integration and a consistent default plugin set matter more than binary-size control.
• The bundled AI providers and built-in stabilization implementation match product policy.
• A single application-level owner can accept global logging and audio-session setup.

When to Prefer Modular Integration

• Only a small subset of device features is needed.
• The app supplies its own AI stack or optional media processors.
• Binary size, third-party transitive dependencies, or startup side effects require tight control.
• The app already owns a sophisticated audio-session policy.
• Security policy excludes diagnostic servers or crash capture from a target.

Avoid Double Initialization

After All-in-One succeeds, do not separately initialize AIBudsSDK or AIBudsAISDK, and do not register competing implementations for plugins that All-in-One already selected. Feature code should consume the initialized suite rather than create another global owner.

If initialization returns false, do not continue with feature startup. Inspect logs and correct the failed Core or AI stage; the public return value does not identify which dependency failed.

Optional Components Still Need Policy

• Install Crash Reporter early only after defining report consent, retention, and upload behavior.
• Start AI Dashboard only for controlled diagnostic builds and networks.
• Use Live Streaming only when the app owns player/streamer cleanup.
• Keep stabilization source-file fallback and storage policy even though the plugin is pre-registered.

Adoption Checklist

• The product intentionally accepts every All-in-One dependency.
• Initialization occurs once from an application-level owner.
• Core configuration and delegate are finalized before initialization.
• Audio-session behavior is compatible with the host app.
• Crash Reporter and AI Dashboard remain separately policy-gated.
• No duplicate Core, AI, or plugin initialization exists.
• A future switch to modular integration has an owner and regression plan.