← All models
Zonos2 Text to Speech
zonos2
Text to Speechby zonos2
Zonos2 is a text-to-speech model that clones a voice from a short sample and speaks naturally across many languages.
Pricing
- Price
- 1.25 cr / minute
Prices in credits (1 credit = $0.01).
Data schema
Input
| Field | Type | Description |
|---|---|---|
| seed | — | Seed for reproducibility. Random when omitted. |
| text | string | Text to synthesize in the cloned voice. |
| min_p | number | Minimum-probability sampling threshold. |
| top_k | integer | Top-k sampling cutoff (0 disables). |
| top_p | number | Nucleus sampling probability (0 disables). |
| language | string | Text-normalization language code. Supported: en_us, en_gb, fr_fr, de, es, it, pt_br, ja, cmn, ko. Unsupported codes skip normalization. |
| max_tokens | — | Maximum number of audio frames to generate. Defaults to the model context limit. |
| temperature | number | Sampling temperature. |
| accurate_mode | boolean | True = closer voice match; False = more expressive. |
| reference_audio_urlrequired | string | Reference audio to clone the voice from. |
| clean_speaker_background | boolean | Mark the reference audio as having a clean background. |
Output
| Field | Type | Description |
|---|---|---|
| seed | integer | The seed used for generation. |
| audio | — | The generated audio (WAV, 44.1kHz mono). |