Google Gemini 3.8 Flash TTS

Google Gemini 3.8 Flash TTS turns text into natural, expressive speech in 130+ languages, with 30 voices, style control and two-speaker dialogue.

Google's Gemini 3.8 Flash TTS turns text into natural, expressive speech. It is the higher-fidelity of Google's two Gemini 3.8 speech models, suited to narration, scripts with several characters, text with many sound tags and hard-to-pronounce words. It speaks over 130 languages, finds the language from the text itself, and has 30 voices, for example Kore (firm), Puck (upbeat), Zephyr (bright) and Charon (informative). Use it for audiobook and article narration, video voice-overs, podcast and dialogue scenes, voice assistants and read-aloud buttons.

Inputs: text (required: the words to speak, read exactly as written, up to 5,000 characters; short sounds go inside it as tags such as <laugh>, <sigh> or <short pause>), voice (Kore by default, 30 to pick from), style (optional: how to say it, for example cheerful and friendly, whispering or speaking slowly; directions typed into text would be spoken) and speakers (optional: the names and voices of two speakers, such as {"Joe": "Puck", "Jane": "Kore"}; text is then a script with one line for each turn, written Joe: words).

The result is audio, one WAV file (24 kHz, mono, 16-bit). A request makes up to about 10 minutes of speech, and a text over 5,000 characters is refused: send a longer text in parts. It uses the 30 built-in voices: it does not clone voices, take audio in or translate, and it speaks the text in the language it is written in.

Price: Google bills the tokens of the text and of the speech. Until 2026-12-31 text costs $0.50 per 1M tokens and speech $9.00 per 1M tokens; from 2027-01-01 Google doubles both, to $1.00 and $18.00. Speech is 25 tokens a second, so a minute of speech (about 1,000 characters of English: 978 made 61 seconds in our check) costs about $0.0135 until the end of 2026 and about $0.027 after, plus a hundredth of a cent for the text. A refused or failed run costs nothing.

Price: $0.50 per 1M text tokens and $9.00 per 1M audio tokens; speech is 25 audio tokens per second, so about $0.0135 per minute of speech plus the text; one request costs at most $0.150222.

Run Google Gemini 3.8 Flash TTS via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/gemini-3_8-flash-tts/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text":"Welcome to FlyMy.AI. Your order has shipped and will arrive on Friday.","voice":"Kore"}'

API documentation · Create a free account · Pricing

Related models