SUPERMODELS
SUPERMODELS

Google

Gemini 3.8 Flash TTS Text to Speech

Turn scripts into narration or two-speaker dialogue. 30 voices, a style field for delivery, inline sounds and pauses, auto language. Pay per use.

Explore this model No subscription. Sign in to create.

Make your budget tangible

What does $10 make?

All your selected credit on Gemini 3.8 Flash TTS.

$10$500
200estimated script runs

$0.05 per reference run

1,000 input characters; estimated audio at 12 characters/second and 32 audio tokens/second. Settings and source media affect the estimate. Review the studio price before creating.

The practical differences

Compare Gemini 3.8 Flash TTS features and pricing

Supported controls in Supermodels, rather than measured quality rankings. Prices use each model’s stated reference workload.

CapabilitiesGemini 3.8 Flash TTSGoogleGrok Text-to-SpeechxAIEleven Turbo v2.5ElevenLabs
Reference price$0.051,000 input characters; estimated audio at 12 characters/second and 32 audio tokens/second$0.031,000 input characters; default voice settings$0.081,000 input characters; default voice settings
Voice selectionSupportedSupportedSupported
Two-speaker dialogueSupportedNot availableNot available
Language detectionAutomaticNot availableNot available
Language controlNot availableSupportedSupported
Speed controlNot availableNot availableSupported
Output formatMP3 · WAVMp3 · Wav · Pcm · Mulaw · AlawNot available

From reading to making

Explore Gemini 3.8 Flash TTS

Prepare your inputs and inspect the estimate. Generation requires an account and wallet credit.

PROMPT0 / 4,000 characters
VOICE PREVIEW COPY

You're finally here, and I've been waiting longer than you can imagine. You found the old key in the attic, didn't you? I know which door it opens, but before you turn it, there's something you need to understand.

GOOGLE · Lyria 3 Pro

Lyria 3 Pro

Generate full-length songs up to 3 minutes from text prompts or images with Lyria 3 Pro, Google's most capable music generation model

Lyria 3 Pro · 1 output

Where Gemini 3.8 Flash TTS fits

Create a narrated scene, podcast exchange, character dialogue or multilingual voiceover.

A useful starting point

Start with an empty style for natural delivery. For two voices, choose a second voice and write one turn per line as Name: words. Use Name (style): words to direct an individual line.

Know the tradeoffs

Two-speaker mode requires exactly two names. Output duration and token use are estimated from the script, so delivery and pauses can change the audio length.

Supermodels is an independent creative studio. Model capabilities shown here reflect the controls available in this integration.

Gemini 3.8 Flash TTS on Replicate

Gemini 3.8 Flash TTS questions and answers

What can I use Gemini 3.8 Flash TTS for?

Create a narrated scene, podcast exchange, character dialogue or multilingual voiceover. The studio shows the inputs and settings supported by this Supermodels integration. Review the model’s limitations before choosing it for your project.

How much does Gemini 3.8 Flash TTS cost on Supermodels?

There is no subscription. The price depends on the model, settings, and requested output. Use the calculator for a stated reference workload, then review the studio estimate for your inputs. Wallet funding is separate from the amount spent on a run.

Can I try Gemini 3.8 Flash TTS before signing in?

You can explore the supported controls and price estimates before signing in. Generating a result requires an account and wallet credit. Supermodels is an independent service; the controls shown apply to this integration.

Make something.

Free sign up

No subscription. Add credit when you’re ready to create.