Richard Dix Stewart

I’m Richard Dix Stewart, a voice performance professional and AI data specialist blending live theater experience with editorial precision to train natural, expressive English for AI systems. My stage background has sharpened my instincts for pacing, tone, and character, which I apply to structured data creation, annotation, and quality assurance in language model training. I actively contribute to multiple voice AI and language model development platforms, focusing on clear, expressive, and instruction-adherent speech data.

Today I work across several programs to help AI systems understand human speech with nuance and reliability, including unscripted and scripted voice data, multimodal annotations, and brand voice direction. I thrive at the intersection of creativity and rigor, bringing a deep performer’s sensibility and a meticulous editorial mindset to every data contribution and training project.