Speech Synthesis
For fo-elevenlabs, the host sends spoken text and the selected persistent voice binding. The ElevenLabs adapter returns audio for ordinary project-media ingestion and a usage receipt.
fo-fal renders the catalogue’s Eleven v3 speech row on fal from text. Its audio is ordinary project media, added at Voice samples or Sound samples.
An uploaded WAV or MP3 alone is not a reusable ElevenLabs voice ID. Use Voice Generation to design, audition and save a voice first. The ElevenLabs integration produces 24 kHz WAV; streaming and cloning are outside its implemented scope.
Subtypes
Section titled “Subtypes”- Text + provider voice ID to WAV — implemented
- Text to speech on fal (Eleven v3) — implemented by fo-fal