Add gladia-podcast-shorts example

[soniasriv]

Adds an example that cuts a long podcast recording into short vertical clips.

Gladia transcribes the recording once with speaker labels. The script scores every window of speech by length and keyword hits, keeps the best non-overlapping windows, slices an SRT for each from the same transcript, and renders a 9:16 short with captions and the speaker label.

Captions come from the existing transcript, so no clip is transcribed twice.

• Zero dependencies, Node 20, .env.example with empty values
• README follows STANDARDS.md
• Re-runs are resumable through output/manifest.json
• Template passes shotstack validate
• Rendered end to end on the sandbox, with a stock audio file and sample SRT in place of the Gladia output

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated
reelestate	Ignored	Preview	Sep 25, 2026 8:44am UTC