{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/8d8a03b0ec9841db856c3913d0620ae3\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/8d8a03b0ec9841db856c3913d0620ae3-84e18ba6b871f713.gif","duration":212.4876,"title":"Testing Text-to-Speech Model Variations and Performance","description":"In this video, I conducted a quick test comparing the March and December snapshots of the OpenAI text-to-speech model, focusing on how well it follows different instructions. I tested various styles, including high-pitch, low-pitch, whispering, and aggressive accents, and found that both versions performed similarly. Despite trying different voices, I couldn't notice any significant differences in the December snapshot when using the instructions parameter. I would appreciate any feedback on this, as I'm unsure if I'm missing something or if there's a different approach I should take. Cheers!"}