WORKBENCH / SPEECH GENERATION

Gemini TTS TEXT → AUDIO

Turn text into speech and create a conversation between two speakers.

YOUR KEY. YOUR EXPERIMENT.

01

Speech settings

Get a Google API key ↗

API key PRIVATERemember API key on this device

Keys stay in memory by default. Speech generation costs are billed to your Google account.

Modelgemini-3.8-flash-ttsSpeaker modeVoiceDelivery style OPTIONAL

Enter ongoing emotion and pacing instructions separately. Put only the spoken dialogue in the text field.

02

Text to read

SINGLE-SPEAKER

Enter the text to turn into speech.

Requests are sent to Google Gemini TTS.

03

Audio results

AWAITING INPUT

♫

Where text finds its voice

Set your dialogue and voices.
Generated audio will appear here.

NO AUDIO YET

EXPLORE / HOW TO USE

Gemini text-to-speech playground

Create speech with Gemini TTS using your own Google API key. Choose voices and delivery style, create single- or two-speaker audio, and download WAV files.

What you can try

Choose a TTS model and built-in voices.
Create narration or a dialogue between two speakers.
Set delivery style, play generated audio, and download a WAV file.
Requests use your provider’s API quota and may incur charges. Check the key, model access, and billing if you receive an authentication or quota error.

How to get started

Enter a Google API key and choose the model and speaker mode.
Select voices and write narration or dialogue lines using Speaker 1: and Speaker 2:.
Generate the speech, listen to it, and download the WAV file.
Example: generate a short welcome message, then change the voice or delivery style while keeping the text fixed.

Explore all playgrounds →API key guide →