PicX Studio

AUDIOGoogle · from 10 credits / generation

Gemini 3.1 Flash TTS, directed from the script

Gemini TTS: Text to Speech Generator

Inline audio tags. 70+ languages. Up to 10,000 characters. Read in Kore.

Google built Gemini 3.1 Flash TTS to be steered from inside the text: drop a tag like [whispers] or [excited] before a line and the read changes there. PicX reads every clip in Kore, the voice Google sums up as firm.

Try Gemini TTSSee examples
Artwork: an illustration for this page, not model output
Artwork: an illustration for this page, not model output
  • Text

Direct the read inside your script

Audio tags change delivery mid-sentence

The headline of the 3.1 Flash release is control from the text itself. Put a bracketed cue such as [whispers], [shouting], [pensive] or [caution] right before the words it should colour, and the voice shifts at that point, then carries on. Google publishes no closed list, so plain-English cues are worth trying; keep them about the voice, not sound effects.

Kore: one steady voice for every clip

Google gives each of its 30 prebuilt voices a one-word character, and Kore's is firm. PicX reads every clip in Kore, so a series of takes sounds like the same speaker. Choosing one of the other 29 voices, such as upbeat Puck or warm Sulafat, is not on PicX yet.

Write in your audience's language

Google lists 70+ languages for 3.1 Flash TTS. Paste the script in the language you need and keep the tags in English; there is no language switch to set.

Where an expressive voice earns its keep

  • App and product voices Weather updates, onboarding lines and alerts that sound glad, calm or urgent where the copy needs it.

  • Audio guides Museum, tour and walkthrough narration, with a hushed aside or a pause written right into the text.

  • Story and character reads One narrator voicing a scene: switch from a whisper to a shout between lines with tags.

  • Localised voice-overs The same explainer read in Spanish, German or Japanese, all in the same Kore voice.

  • Lip-synced video Download the clip and use it as the new audio track in AI Lip Sync.

Gemini TTS

Why pick Gemini 3.1 Flash TTS

Try Gemini TTS
  • [tags]Inline audio tagsSteer tone and pace per line.
  • KoreOne consistent voiceGoogle's firm voice on every take.
  • 70+LanguagesPer Google's launch post.
  • 10KCharacters per clipLong scripts in one take.
  • 1,211Elo at launchArtificial Analysis TTS leaderboard, per Google.
  • SynthIDWatermarked audioGoogle marks every clip as AI-made.

What Gemini TTS costs

One flat price per generation. The same price the generator shows before you run it.

  • Minimum per generation
    10credits / generation
    On Starter≈ 360 generations / mo
  • Per 1,000 characters
    50credits / generation
    On Starter≈ 72 generations / mo

Every model on every planCredits never expireSee plans

Pay for Gemini TTS with crypto

Buy credits with Bitcoin, USDT, USDC, ETH, Solana and more — no card, nothing auto-renews, packs from $5, and 15% extra credits on crypto payments.

How crypto checkout works

Gemini 3.1 Flash TTS vs Gemini 3.8 Flash TTS

PicX runs Gemini 3.1 Flash TTS. Google's newer 3.8 Flash TTS adds custom voices, but it is not on PicX yet.

Gemini 3.1 Flash TTS (on PicX)Gemini 3.8 Flash TTS (not on PicX yet)
Prebuilt voices30 in Google's API; PicX reads in KoreThe same 30 prebuilt voices
Design or clone a voiceNoYes, in Google's API
Inline delivery cuesBracketed tags, e.g. [whispers]Angle-bracket tags, e.g. <whispers>
Languages (per Google)70+130+
Speakers per clipOne on PicX (the model can do two)Up to two with prebuilt voices

Live credit prices for Gemini TTS are in the pricing table.

How to use Gemini TTS in 3 steps

  1. 1

    Write the script, with cues

    Paste the exact words to be spoken. Add a bracketed tag such as [whispers] or [excited] before any line that needs a different delivery.

  2. 2

    Check language and length

    Write in the language you need, with the tags in English, and keep it under 10,000 characters. Each clip is read by one voice, Kore.

  3. 3

    Generate and download

    Listen back, move or swap a tag if a line lands wrong, and download the take from the generator or My media.

Gemini TTS prompts to copy

Prompts to start from. Copy one, or open it in the generator with the model already picked.

  • Weather report with a mood shift

    [excited] Good morning! The sun is finally back this afternoon, with a light breeze off the coast. [caution] Keep an umbrella close tonight, though... showers roll in after dark.

    Try it

  • Hushed museum guide

    [whispers] Step a little closer. [pensive] This is the oldest painting in the collection, finished in fifteen eighty and still in its original frame. Notice how the old varnish has turned the sky almost gold.

    Try it

  • One narrator, two characters

    The gate groaned open, and the traveller froze. [shouting] Halt! State your name and your business! [whispers] Please... I only carry a letter for the elder.

    Try it

  • Spanish app onboarding

    [excited] ¡Bienvenido! Toca el botón más para empezar tu primer proyecto. Nosotros guardamos tu trabajo mientras avanzas.

    Try it

Gemini TTS questions

Which Gemini TTS model does PicX Studio use?

Gemini 3.1 Flash TTS, the version Google released in April 2026 with inline audio tags. Gemini 3.8 Flash TTS, 3.8 Flash-Lite TTS and 2.5 Pro TTS are not on PicX yet.

Is Gemini TTS free?

Not on PicX Studio. Your first image is free after you sign up, but Gemini TTS uses credits, charged by the length of your text. Credits never expire and work on every model and tool.

What voices does Gemini TTS have?

Thirty prebuilt voices, each with a one-word character from Google: Zephyr (bright), Puck (upbeat), Charon (informative), Kore (firm), Fenrir (excitable), Leda (youthful), Orus (firm), Aoede (breezy), Callirrhoe (easy-going), Autonoe (bright), Enceladus (breathy), Iapetus (clear), Umbriel (easy-going), Algieba (smooth), Despina (smooth), Erinome (clear), Algenib (gravelly), Rasalgethi (informative), Laomedeia (upbeat), Achernar (soft), Alnilam (firm), Schedar (even), Gacrux (mature), Pulcherrima (forward), Achird (friendly), Zubenelgenubi (casual), Vindemiatrix (gentle), Sadachbia (lively), Sadaltager (knowledgeable) and Sulafat (warm). PicX reads every clip in Kore for now; choosing one of the other 29 is not on PicX yet.

How do I use Gemini TTS audio tags?

Write a short cue in square brackets right before the words it should affect, for example "[whispers] Step a little closer." or "[shouting] Halt!". The delivery changes at the tag and you can switch again later in the same sentence. Google does not publish a fixed list, so try plain cues like [excited], [pensive] or [caution], and write them in English even when the script is in another language.

Gemini TTS vs ElevenLabs: which should I use?

Both are on PicX Studio and each reads in one preset voice here. Reach for Gemini TTS when you want to steer the read line by line with bracketed tags, or need one of the 70+ languages Google lists for it. Run the same short script through both and keep the read you prefer before starting a long job.

Can Gemini TTS do two speakers or clone a voice?

Google's model supports two-speaker dialogue, and Gemini 3.8 adds voice design and cloning, but neither is on PicX yet. Each PicX clip has one preset voice. For a dialogue, generate one clip per voice and edit them together.

What languages does Gemini 3.1 Flash TTS speak?

Google lists more than 70 languages for 3.1 Flash TTS. Write the script in the language you want; there is nothing to switch. Accent controls are not offered on PicX.

How much does Gemini TTS cost?

It is billed in PicX credits by the length of your text, up to 10,000 characters per clip. The pricing table on this page shows the live price, read from the same catalog the generator charges from.

Give your script a voice, and a mood

Google's expressive TTS, cues written into the text, pay by length.

FREEClaim Free Image Generation