{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/e9aa2df96be3484ab852f04cac68e46b\" frameborder=\"0\" width=\"1150\" height=\"862\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":862,"width":1150,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":862,"thumbnail_width":1150,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/e9aa2df96be3484ab852f04cac68e46b-aafa38e21ce28203.gif","duration":868.377,"title":"How Sri Studio Automates Reel Creation","description":"This Loom explains why Sree Studio was built as an end-to-end pipeline for generating vertical launch reels from a single sentence and a voice clone. The creator describes choosing option three over simpler alternatives, implementing a five-step Python pipeline with a mandatory human approval gate, parallel execution for voice cloning, image generation, caption timing, and video stitching, plus a final brand-voice eval via Langfuse. They discuss tool choices, including rejecting N8n in favor of a Python-native LangRaph approach for shared state, retries, and simpler maintenance, with overall runtime of about 8 to 10 minutes and typical cost around 25 to 30 cents. The Loom also covers testing, observability, and failure handling to ensure the reel ships reliably."}