Description

KeenResearch AI specializes in offline speech recognition technology. Their SDK powers on-device speech recognition across mobile platforms, processing audio locally without any server dependency.

As a software engineer at KeenResearch, my primary role was porting their C++ SDK to WebAssembly, bringing these capabilities to the browser for the first time. This opened up an entirely new market for the company, expanding their product offering beyond mobile into web.

My work can be found on:

• Demos: https://keenresearch.com/demos
• Web SDK Docs I've written: https://keenresearch.com/keenasr-docs/keenasr-web-introduction.html

Responsibilities

• C++ to WebAssembly port — Ported the core speech recognition SDK to WebAssembly using Emscripten and Node.js, enabling offline speech recognition directly in web browsers.
• SDK documentation — Authored comprehensive documentation with implementation guides and API references, making the SDK accessible to external developers.
• React demo applications — Built interactive demo apps using React and Web Audio API to showcase the SDK's capabilities and performance to potential customers.
• Performance optimization — Optimized memory management and audio processing for cross-browser compatibility and efficient runtime execution.