{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/888127ff903f437bb557b598fcc2614b\" frameborder=\"0\" width=\"1418\" height=\"1063\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1063,"width":1418,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1063,"thumbnail_width":1418,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/888127ff903f437bb557b598fcc2614b-5ca8d3ef3d7bdbd4.gif","duration":353.285,"title":"DeepMedia's GenAI Voice Benchmark Dataset Showcase","description":"Hi, I'm Rijal Gupta, the founder and CEO of DeepMedia. Today, I'm showcasing our latest GenAI voice benchmark dataset, the largest and most diverse for public benchmarking. If you're in research or tech, fill out the form for access. The dataset includes 16 generative capabilities and 8,000 speaker identities. It's crucial for ensuring detectors learn from AI artifacts, not just identities."}