30 VOICES · SINGLE OR TWO SPEAKERS
Text to Speech Online, Powered by Gemini 3.8 TTS.
Create narration or dialogue with Gemini TTS. New accounts get 20 free credits.
REAL AUDIO · GENERATED WITH GEMINI 3.8
Hear the difference.
Listen to a single voice, a Mandarin reading, and a two-speaker exchange. These clips were generated for this site and are served from our own media storage.
SCRIPT EXCERPT
“Hello! Welcome to Gemini TTS.”
What is Gemini TTS?
Gemini TTS turns a written script into speech you can play in your browser and download as WAV. Use one voice for a reading or assign lines to two speakers for a conversation. The studio offers Gemini 3.8 Flash and Flash Lite, 30 preset voices, and controls for accent, style, and pace. Add short cues such as <laugh> to the script, or describe the voice and scene in advanced controls. Preview a voice before generating, then find finished audio in your library.
Try it above: choose a model and voice, enter up to 5,000 characters, and check the estimated credits before generating.
How to create speech with Gemini TTS.
The Gemini TTS generator is above this guide. Start with a short script, then refine the delivery before spending credits.
Choose a model and voice
Use Flash for expressive narration or compare Flash Lite at its lower credit rate. Browse 30 voices and play a neutral sample. For dialogue, choose a voice for each speaker.
Write and direct the script
Type or paste your text, then adjust accent, style, and pace. Add <sigh> or <short pause> where needed. Optional voice and scene descriptions give the model more context.
Generate and keep the take
Check the estimated credits, then generate. Play the result in the studio or download its WAV file. Completed Gemini TTS takes remain in your library.
Make the voice fit the story.
A clear script and a few choices about the speaker and delivery can shape the result.
Video narration
Turn an explainer or product walkthrough into a voiceover. Pick one voice, adjust the pace for your edit, and listen before downloading the WAV file.
Explore workflowStories and read-aloud drafts
Hear a passage aloud. Try another voice or style to check whether a scene feels calm, urgent, or conversational.
Explore workflowTwo-speaker dialogue
Assign each line to speaker one or two. Choose contrasting voices, add pauses or vocal cues, and generate the exchange as one audio result.
Explore workflowTwo-host podcast
Start with a written episode script, divide it between two hosts, and download the recorded exchange for editing. Use an opening, interview, or explanation template.
Explore workflowChoose Flash or Flash Lite for your script.
Both models support single-voice and two-speaker scripts here. Choose for the task, then try the same script with each model to compare results.
Gemini 3.8 Flash
Choose Flash for a narrated story, character dialogue, or changing tone. Direct the performance with style, pace, and cues in your text.
Gemini 3.8 Flash Lite
Choose Gemini TTS Flash Lite for straightforward read-aloud audio or repeated drafts. Select a voice and compare its result with Flash on the same line.
FIND YOUR NARRATOR
Listen before you choose.
Browse all thirty Gemini TTS voices in one place. Each card has a neutral audio sample, a short description, and a direct link to use that voice in the studio. Search by name or explore a delivery character such as warm, steady, or conversational. Listening to an existing preview does not submit a generation request.
Choose a small shortlist, then compare the same passage with each candidate. Include a name, number, or phrase from your actual project so the test tells you something useful. A voice that works for a greeting may need different direction for a technical explanation or a story. Keep the script and model unchanged while you compare.
For a two-person exchange, listen to the pair together. Give each participant a clear role, assign the lines in order, and check that a listener can follow who is speaking. The dialogue templates offer an interview, a lesson, and a short story scene as starting points.
WRITE FOR THE EAR
Give the voice a clear direction.
Google Gemini text to speech models provide the speech capabilities behind this independent application. Your script still does much of the work: shorter sentences, familiar wording, and a clear sequence make an explanation easier to hear. Start with the words, then choose the delivery that supports their meaning.
In Gemini TTS, keep spoken content in the script and place general instructions in the voice description or scene field. A short direction such as “patient teacher, clear and conversational” gives you a useful first test. Add a supported pause or vocal cue only where the moment calls for it, then listen before adding another.
The Gemini TTS prompting guide includes examples you can copy or apply in the studio. Templates remain editable and do not automatically start a task. Review the selected model, character count, and credit estimate before generating. Download an approved WAV take for your project, and keep your script and settings as a reference for the next segment.
Before you press generate.
What can I make?+
Create narration, read-aloud audio, and two-speaker dialogue. Gemini TTS lets you choose a voice for each speaker.
Which model should I choose?+
Start with Gemini 3.8 Flash for expressive delivery or dialogue. Try Flash Lite at its lower credit rate for read-aloud work or repeated drafts.
Can I add pauses or other vocal cues?+
Yes. Use the buttons below the script to insert <laugh>, <sigh>, or <short pause> where you want the delivery to change.
How much text can I enter?+
Each generation accepts up to 5,000 characters. In two-speaker mode, the combined lines count toward that limit.
Can I download the result?+
Yes. Play or download completed audio as WAV from the studio. Finished takes remain in your library.
How much does it cost to try?+
New accounts receive 20 starter credits. The estimate beside the generate button reflects the selected model and script length. Sign in to see your balance, and review Pricing if you need more.
Is this a Google product?+
No. Gemini TTS is an independent application built around Gemini speech models; it is not a Google product.
The next voice starts with your words.
Open the studio, write a line, and hear it come to life.
Open the studio