Header Logo

Docs

Start

Header Logo

Start

Header Logo

Docs

Start

All Models

WaveSpeed logo

Zonos2

Zonos2 turns written text into speech using a voice cloned from reference audio, supporting speaker-specific narration and dialogue prototyping.

Text to speech

Based on verified input duration.

OfferingPrice (USD)
All listed options$0.01 / run
  • Clean Speaker Background: Off, On (same price).
  • Based on the audio input; billing steps apply in this order:
  • Round input duration up to a whole second.
  • Cap billed duration at 30 seconds.
  • Convert to minutes.
  • Round input duration up to a whole minute.
  • Bill at least 1 minute.
  • Round the total charge up to the next $0.01.

Zonos2

Zonos2 is a text-to-speech model from Zyphra that uses reference audio to clone a speaker's voice. The audio supplies the voice identity, and the written text supplies the words to be spoken. The result is generated speech in the referenced voice.

This pairing is useful when a production has a chosen speaker and needs to hear additional written material in that voice. Scripts can be revised independently of the reference recording, making it possible to explore alternate wording or develop narration as an edit takes shape.

What it's good for

Video editors can create narration for draft cuts, and writers can audition dialogue variations during script development. Product teams can also generate spoken material for demos or app experiences using a target voice. For audio-led creative work, the reference recording gives speech generation a specific vocal starting point tied to the project.

Model ID

wavespeed-ai/zonos2

Output Type

Audio

Pricing

$0.01 / run

Provider

WaveSpeed

Hosting

Protoface

Zonos2

Zonos2 is a text-to-speech model from Zyphra that uses reference audio to clone a speaker's voice. The audio supplies the voice identity, and the written text supplies the words to be spoken. The result is generated speech in the referenced voice.

This pairing is useful when a production has a chosen speaker and needs to hear additional written material in that voice. Scripts can be revised independently of the reference recording, making it possible to explore alternate wording or develop narration as an edit takes shape.

What it's good for

Video editors can create narration for draft cuts, and writers can audition dialogue variations during script development. Product teams can also generate spoken material for demos or app experiences using a target voice. For audio-led creative work, the reference recording gives speech generation a specific vocal starting point tied to the project.