Google Gemini 3.1 Flash TTS

Google Gemini 3.1 Flash TTS is a new text-to-speech model that makes AI-generated speech sound better. It helps create audio that is more natural and interesting in more than 70 languages. This model is designed to improve the quality and expressiveness of AI voices.

Benefits

Gemini 3.1 Flash TTS offers more control over how AI voices sound. It uses special audio tags that let you guide the voice's style, speed, and how it's spoken using simple text commands. This means you can fine-tune voices, set the mood, and give specific directions for pace, tone, and accent. This makes it easier to create consistent and high-quality voiceovers for various projects.

Use Cases

Developers can use Gemini 3.1 Flash TTS with tools like Google AI Studio. They can adjust voices and add specific instructions for delivery. The settings can then be saved as code for the Gemini API, ensuring the voice sounds the same every time it's used. This model is available to developers through the Gemini API and to businesses on Vertex AI. It will also be available to Google Workspace users through Google Vids.

Additional Information

To make sure the technology is used responsibly and to prevent the spread of fake information, all audio created with Gemini 3.1 Flash TTS will have an invisible watermark. This watermark, created by SynthID, can show that the audio was generated by AI.