falText to speechComing soon

fal-ai/zonos2

Name: fal-ai/zonos2
Brand: fal

fal-ai/zonos2 is a text to speech tool from fal that runs in the background and returns audio. Coming soon: listed with its price, not yet open to runs.

This tool is listed but not yet available to run.

Price pending

Pricing

Price pending.

Limits

Runs: In the background
Returns: audio
Longest queue wait: 3600s
Run deadline: 1800s
Largest inline input: 1024 KB
Largest input file: 500 MB
Largest output: 4768 MB
Outputs per run: 16

Input

Field	Kind	Default	Description
`reference_audio_url` required	text	—	Reference audio to clone the voice from.
`text`	text	"Fal" is the fastest solution for your audio generation.	Text to synthesize in the cloned voice.
`language`	text	en_us	Text-normalization language code. Supported: en_us, en_gb, fr_fr, de, es, it, pt_br, ja, cmn, ko. Unsupported codes skip normalization.
`seed`	number (whole) 0 to 2147483647	—	Seed for reproducibility. Random when omitted.
`accurate_mode`	true or false	true	True = closer voice match; False = more expressive.
`clean_speaker_background`	true or false	false	Mark the reference audio as having a clean background.
`temperature`	number 0 to 2	1.15	Sampling temperature.
`top_p`	number 0 to 1	0	Nucleus sampling probability (0 disables).
`min_p`	number 0 to 1	0.18	Minimum-probability sampling threshold.
`top_k`	number (whole) 0 to 2048	106	Top-k sampling cutoff (0 disables).
`max_tokens`	number (whole) 1 to 6144	—	Maximum number of audio frames to generate. Defaults to the model context limit.

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/zonos2",
    "input": {
      "reference_audio_url": "https://storage.googleapis.com/falserverless/model_tests/zonos/demo_voice_zonos.wav"
    }
  }'