AI podcast pipeline: GPT-Image, LTX, ElevenLabs and Soniox generate two-host shows
sven_ai · x · 2026-10-20
A multi-model pipeline demonstrates fully AI-generated two-host podcast videos: GPT-Image generates high-aesthetic studio scenes and consistent characters; LTX drives micro-expressions and body motion; ElevenLabs + Gemini blend dual-voice audio; Soniox provides millisecond-level timestamps, and Opus + HyperFrames handles shot cutting and packaging entirely in code — no After Effects.
A directly reusable stack for anyone building virtual hosts or AI content pipelines.
More from coding & agent
- Lawyers are Claude's #1 power users as legal usage grows 108x — how to build your own legal team with it — rubenhassid · 2026-10-20
- Dev installs a Mac-connected terminal on a Samsung Frame TV, eyes porting Claude Code next — CtrlAltDwayne · 2026-10-20
- RAG reranking shootout: Cohere lifts an answer from rank 15 to 1 while a small cross-encoder makes it worse — Affectionate_Slip654 · 2026-10-20
- Context.dev raises $9M led by General Catalyst to give AI agents one API for web data — rohanpaul_ai · 2026-10-20
- Indie dev says AI removed the model bottleneck: solo-building a physics excavator sim — Michael_J_Black · 2026-10-20
- Replit Hits #2 on Slides Arena with Elo 1279, Beating Gamma and Manus — amasad · 2026-10-20