Generate audio
Submits an asynchronous audio (text-to-speech) generation job to the fal.ai queue and returns the queued job immediately. It speaks the given text; for a song, soundtrack or instrumental use POST /v1/audio/music. Poll GET /v1/jobs/{id} or GET /v1/audio/{id} for status and results. Credits are charged only when the job completes successfully.
Authorizations
API key in the format: Bearer {api_token}
Headers
Makes this write safe to retry: a repeat with the same key replays the first response instead of creating a second resource. Scoped to the account, remembered for 24 hours; reusing a key with a different payload is an error.
255Body
Model-specific input, forwarded to fal verbatim. text is required.
A quality tier (fast, balanced, quality) or any fal.ai text-to-speech endpoint id. Prefer a tier — it tracks the best current model for that price point. Defaults to fal-ai/elevenlabs/tts/multilingual-v2; GET /v1/audio/models maps tiers to ids with live pricing.
"fast"
"balanced"
"quality"
"fal-ai/inworld-tts"
"fal-ai/elevenlabs/tts/turbo-v2.5"
"alibaba/qwen-audio-3-tts"
"fal-ai/elevenlabs/tts/multilingual-v2"
"fal-ai/elevenlabs/tts/eleven-v3"
Response
Generation job queued
queued, processing, completed, failed, cancelled 
