At RWS TrainAI – Project Echo, I evaluate Indonesian AI-generated speech for accuracy, fluency, pronunciation, tone, and naturalness. I assessed 100+ audio samples, comparing outputs and recording judgments against project guidelines.
Across CrowdGen, Mercor, and Babel Audio, I annotate Indonesian-language data, evaluate content, and correct transcripts against original audio. Before this work, I taught and assessed Indonesian language skills at Pondok Pesantren Zaadul Ma’ad, including grammar, reading, writing, and punctuation.

