Workflows
AAD — Audio A Day
Daily TTS audio, transcribable + searchable.
Spoken renditions of PAD pages and LAD lists — TTS via Voxtral / ElevenLabs / OpenAI / local, Whisper transcripts for search. Browsable in the AUDIO gallery.
- Multi-provider TTS routing
- Whisper VTT transcripts
- Searchable audio archive
Why
Text gets read once. Audio gets played in the car, the kitchen, the gym. AAD doubles the lifespan of every PAD or LAD by spawning a spoken version with transcript — searchable, indexable, shareable.
How
- Render TTS for each PAD page / LAD list
- Provider routing: Voxtral · ElevenLabs · OpenAI · local
- Whisper transcripts for search + captions
Proof
- Audio hours archived
- 200+
- TTS providers wired
- 4
- Searchable
- every minute
AAD — audio a day, transcribed and searchable
Text in · TTS routed · transcript indexed
Hover or tap a node to see details.
FAQ
- How do you pick a TTS voice for each piece?
- The router maps content type to voice persona — narration voice for PAD pages, ranking voice for LAD lists. Consistency per stream over time, not per generation.
- Why bother transcribing AI-generated TTS back to text?
- VTT transcripts make every minute searchable, captionable, and re-usable. The original Markdown is for reading; the VTT is for indexing and accessibility — two different jobs.
- Can the audio archive be served as a podcast?
- The archive auto-builds an RSS feed compatible with Apple Podcasts, Spotify, and Pocket Casts. Same render, no extra work — every PAD/LAD becomes podcast-discoverable.
In production
- AUDIOAI catalogue
200+ hours of TTS audio with Whisper VTT transcripts — every minute searchable.
See it - PAD/LAD spoken renditions
Every daily text artefact gets a spoken version for free — same content, different lifespan.
- Four-provider TTS router
Voxtral, ElevenLabs, OpenAI, local — picked per content type and voice persona.