Independent AI Evaluation & Content Creation Projects (Sora, Kling, Sundance)
Conducted over 1,000 structured AI evaluation experiments comparing generative audio outputs between English and Spanish prompts. Detailed documentation of differences in tone, emotional delivery, continuity, and naturalness of responses. Applied nuanced assessments for timing, rhythm, prosody, and non-verbal cues in multilingual outputs. • Performed bilingual evaluation for accent stability and speech naturalness in generated audio samples. • Conducted repeated testing and iterative corrections to refine output quality and variability. • Built multi-shot audio sequences to showcase AI advancements, including public portfolio demonstrations. • Provided high-level feedback on emotional tone and delivery for use in AI audio model improvement.