Newsroom features
Turn your newsroom's articles
into a daily audio edition.
Connect your feed, set a schedule, and every edition is fetched, scripted, rendered, and published — hands-free, with editorial control where it matters.
Core features
Newsroom automation
Connect RSS, WordPress, or any JSON API. Schedule daily or weekly runs. Articles auto-fetch, dedup, generate, render, publish — all hands-free.
RSS distribution
Built-in podcast feed compatible with Spotify, Apple Podcasts, Amazon Music. Submit your URL once — every new episode publishes automatically.
AI script generation
Frontier AI models turn content into multi-anchor dialogue. Batched for long documents. Topic picker for large content so you control what's in.
Multi-anchor episodes
Up to 4 anchors per episode. Persona snapshots carry voice + speed + delivery instructions. Mix and match across templates.
Multi-language voices
English, Hindi, Spanish, French, German, Portuguese, Arabic, Japanese, Korean, Mandarin and more.
73 AI voices
Premium and standard voices across multiple AI engines. Preview before you pick. Voice instructions per anchor for tone control.
100+ templates
Built-in recipes for news, interview, comedy, education, business, kids, meditation, true crime, and more. Each with cover art, music, and persona presets.
83-track music library
Background beds and transition stingers across 13 genres. All commercially licensed (CC-BY 4.0 + Pixabay). Mix volumes per template.
AI cover art
Generate project covers and persona avatars with frontier image models. Upload your own or remix the AI's.
Built for production
Async by default. Workers retry automatically. Failures show up where you can act on them, not buried in logs.
Async generation
TTS workers run in parallel — a 30-segment episode renders in under 2 minutes. Long jobs never block the UI.
Quality gates
Abrupt-cut detection, optional Whisper verification, automatic safety-filter retry. Bad takes get regenerated automatically.
Pronunciation control
Workspace and project-level pronunciation rules. Drop in 'Nguyen → Win' or 'Kubernetes → koo-ber-net-ees' and every TTS call respects it.