You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.

This is a speech synthesis model: you send text, and you get ready-made voiceover as output. It is suitable for video voiceovers, narrating training courses and presentations, voice notifications, answering machines and IVR, audio versions of articles, as well as for prototypes of voice interfaces and assistants.

The model's context is 4095 tokens, output limit is 4096, so it is more convenient to voice long materials in parts and then stitch them together. You get an audio track as output: it can be listened to and downloaded in the BotHub interface or retrieved via API and immediately sent for editing.

A separate reasoning mode is not noted in our data, and it is not required for speech synthesis: the model turns ready-made text into sound. If lines still need to be created or edited, connect a text model in the same BotHub window, and leave the voiceover to WaveNet.

In our data, function calling and structured JSON output are not noted for this model — its result is audio, not a text structure. For scenarios with functions, use a text model: the unified OpenAI-compatible BotHub API allows you to switch without rewriting the integration.

Input file reception is not noted in our data: the working scenario here is text as input, audio as output. If you need to recognize speech from a recording or analyze an image, select a suitable model in the same interface: more than 250 models are available via a single key.

The BotHub interface, payment in rubles, and support are in Russian; access from Russia works without a VPN or foreign card. How well a specific voice fits your material is best checked on a short fragment: there is no subscription, you pay as you go, and tokens do not expire.