Gemini Text-to-Speech vs PATINA

Gemini Text-to-Speech has no write-up on Specifiction yet; its figures are above.

PATINA is a paid API and web service for generating material textures from text or images. Text prompts produce complete, seamlessly tiling physically based materials, with upscaling available to 8192 pixels. The endpoint can output basecolor, normal, roughness, metalness, and height maps. An image URL can be used to make texture variations, while an input image and mask support regeneration of white mask regions while preserving black areas. Tiling may run in both directions, horizontally, or vertically, with tile size and stride controls. Requests support one to four texture images and one to eight denoising steps. Exports are listed as PNG, JPEG, and WebP, and the model listing permits commercial use. Prompt expansion is enabled by default and adds approximately $0.0025. PATINA also has a material-extraction workflow that isolates and flattens material from a photo, makes it seamless, and predicts PBR maps. API examples are provided for JavaScript, Python, and cURL, using a FAL_KEY environment variable.