Gemini 2.5 Flash Preview TTS
Neural Network
Specifications
- Input
- text, images, documents
- Output
- text, audio
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works Gemini 2.5 Flash Preview TTS?
Frequently asked questions about Gemini 2.5 Flash Preview TTS
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 Google model for voice synthesis: it generates audio from text descriptions. It is suitable for voiceovers for videos and educational materials, podcasts, notifications, and voice interface prototypes. It accepts text, images, and documents as input, and can call functions and reason before answering.
Audio generation in addition to text is a confirmed capability, but we will not specify codecs, sample rates, or bitrates: we do not have that data. Refer to the limits in the technical header: 200,000 tokens of context and up to 32,000 output tokens per request.
Yes, the model's reasoning mode is confirmed. This helps to more accurately parse long texts and voiceover scripts: considering the structure of lines, placing accents and pauses according to meaning, and following multi-step instructions. Reasoning consumes output tokens, so keep the 32,000 limit in mind.
Yes, function calling and structured JSON output are supported. The model can be embedded into a pipeline where it calls your tools and returns a predictable schema for further processing. In BotHub, it connects via a unified OpenAI-compatible API, so changing the model does not require rewriting the integration.
According to our data, in addition to text, images and documents are accepted as input — for example, a script in PDF or a layout from a slide. We have no notes on audio and video input, so we will not make any claims: it is easier to check with your own task directly in the interface.
We do not have data on prompt and output languages, so we will not promise specific quality for Russian — it is more reliable to test it with your own texts. In BotHub, you can do this immediately: Russian-language interface, payment with Russian cards in rubles, and access without a VPN or foreign card.