What it does This repo is the Cordova wrapper for Alan AI’s voice platform. You embed it, add a microphone button, and users can talk to your app. The heavy lifting—speech recognition, natural language parsing, and command execution—happens on Alan’s cloud. The SDK itself is mostly plumbing to get audio from the device to their service and route responses back.

The interesting bit Alan AI’s pitch is “Application-Level AI”: their system allegedly ingests your app’s APIs, docs, and GUI structure, then generates features on demand at runtime. That’s a big claim, and this Cordova repo doesn’t demonstrate it—it’s just the client-side microphone hook. The actual magic, if it exists, is locked in their proprietary studio and cloud backend.

Key highlights

• Supports Android via Cordova (iOS isn’t mentioned for this SDK; they have a separate iOS SDK)
• Dialog scripts are written in JavaScript inside Alan AI Studio, not in your app codebase
• Includes example apps with basic voice commands like “Hello” and “What does this app do?”
• Also available for Web, Flutter, React Native, Ionic, and even PowerApps
• Distributed as an npm package: @alan-ai/cordova-plugin-alan-voice

Caveats

• The README is heavy on marketing language (“self-coding engine,” “no developers needed”) and light on technical specifics about how the AI actually generates code
• Requires signing up for Alan AI Studio and relying on their proprietary cloud service
• The “Three-Layer AI (3LAI)” architecture is mentioned but not explained in any verifiable detail

Verdict Worth a look if you’re already committed to Cordova and want voice control without building ASR/NLP yourself. Skip it if you need offline functionality, want to own your AI stack, or are skeptical of platforms where the most impressive claims are also the least inspectable.