Gemini TTS

Google’s Gemini TTS converts text to realistic audio. 30 voice presets, multi-speaker synthesis (up to 10 speakers), 24+ languages, and inline style markers for expressive control.

Pricing: $10/1m-tokens. Endpoint: POST /v1/models/google/gemini-tts/runs.

Google’s Gemini TTS converts text to realistic audio. 30 voice presets, multi-speaker synthesis (up to 10 speakers), 24+ languages, and inline style markers for expressive control.

Overview

• Endpoint: https://api.runflow.io/v1/models/google/gemini-tts/runs
• Model ID: google/gemini-tts
• Provider: Google
• License: commercial
• Last Updated: 2026-04-08

Pricing

• Base price: $10/1m-tokens
• Note: Per 1M tokens (flash, output-dominated)

Gemini TTS API

Endpoint: POST /v1/models/google/gemini-tts/runs

Run the model

Python

import requests

response = requests.post(
    "https://api.runflow.io/v1/models/google/gemini-tts/runs",
    headers={"Authorization": "Bearer RUNFLOW_API_KEY"},
    json={
        "input": {},
        "callback_url": "https://your-server.com/webhook"
    },
)

data = response.json()
print(data)

Node.js

const response = await fetch(
  "https://api.runflow.io/v1/models/google/gemini-tts/runs",
  {
    method: "POST",
    headers: {
      "Authorization": "Bearer RUNFLOW_API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
        "input": {},
        "callback_url": "https://your-server.com/webhook"
    }),
  }
);

const data = await response.json();
console.log(data);

cURL

curl -X POST https://api.runflow.io/v1/models/google/gemini-tts/runs \
  -H "Authorization: Bearer $RUNFLOW_API_KEY" \
  -H "Content-Type: application/json" \
  --data-binary @- <<'JSON'
{
    "input": {},
    "callback_url": "https://your-server.com/webhook"
}
JSON

Request parameters

Parameter	Type	Required	Description
input	object	required	Model input parameters. See “Input schema” below.
callback_url	string | null	optional	Webhook URL - POSTed when the run reaches a terminal state.
metadata	object | null	optional	Arbitrary key-value pairs attached to the run.

Input schema

Field	Type	Required	Allowed values	Description
prompt	string	required	Any	The text to convert to speech. Gemini TTS supports natural-language prompting for style, pace, accent, and emotional expression - include delivery instructions inline with the text (e.g. ‘Say cheerfully: Have a wonderful day!’). For multi-speaker synthesis, prefix lines with speaker aliases defined in the speakers field (e.g. ‘Alice: Hello!\nBob: Hi!’). Supports inline pace/style markers like [slowly], [whispering], [excited], [extremely fast].
speakers	json	optional	Any	Multi-speaker voice configuration. When set, enables multi-speaker synthesis where different parts of the text are spoken by different voices. Each speaker needs a voice and a speaker_id (alias) that matches prefixes in the prompt. Requires gemini-2.5-pro-tts or gemini-2.5-flash-tts model. Not supported with gemini-2.5-flash-lite-preview-tts.
output_format	string	optional	wav, mp3, ogg_opus	Audio output format. mp3: compressed, small file size (recommended). wav: uncompressed PCM wrapped in WAV (24 kHz, 16-bit mono). ogg_opus: Ogg container with Opus codec, good quality-to-size ratio.
model	string	optional	gemini-2.5-flash-tts, gemini-2.5-pro-tts	Which Gemini TTS model to use. gemini-2.5-flash-tts: low latency, cost-efficient for everyday applications (recommended). gemini-2.5-pro-tts: highest quality, best for structured workflows like podcasts, audiobooks, and customer support.
voice	string	optional	Achernar, Achird, Algenib, Algieba, Alnilam, Aoede, Autonoe, Callirrhoe, Charon, Despina, Enceladus, Erinome, Fenrir, Gacrux, Iapetus, Kore, Laomedeia, Leda, Orus, Pulcherrima, Puck, Rasalgethi, Sadachbia, Sadaltager, Schedar, Sulafat, Umbriel, Vindemiatrix, Zephyr, Zubenelgenubi	Voice preset for single-speaker synthesis. 30 distinct voices are available. Ignored when speakers is set. Popular choices: Kore (strong, firm female), Puck (upbeat, lively male), Charon (calm, professional male), Zephyr (bright, clear female), Aoede (warm, melodic female).
language_code	string	optional	Any	Language for multilingual synthesis. When set, steers the model to speak in the specified language. Supports 24 GA languages and 60+ Preview languages. If not set, the model auto-detects the language from the text.
temperature	float	optional	Any	Controls the randomness of the speech output. Higher values produce more creative and varied delivery, while lower values make the output more predictable and focused.
style_instructions	string	optional	Any	Optional style and delivery instructions prepended to the prompt. Controls expressiveness, accent, pace, tone, and emotional expression using natural language. Use this to separate style control from the text content. Examples: ‘Speak warmly and slowly’, ‘Read this as a dramatic newscast’, ‘Use a British accent with a cheerful tone’, ‘Whisper mysteriously’.

Output schema

Field	Type	Description
outputs	json	Unified output array - one entry per generated artifact with url/type/width/height/duration/etc.
nsfw_detected	json	true if the provider flagged output as NSFW, false if cleared, null if not checked.
seed	json	Deterministic seed used for generation, or null if the provider doesn’t return one.
timing	json	Provider timing info (inference_ms etc.), or null.

Callback payload

When you provide a callback_url, Runflow POSTs to it once the run reaches a terminal state.

Field	Type	Description
event	string	Event type: “run.completed”, “run.failed”, or “run.cancelled”.
run_id	string	The unique identifier of the run.
status	string	Terminal status: “succeeded”, “failed”, or “cancelled”.
output	object | null	The run output. Null if the run failed or was cancelled.
duration_ms	number | null	Total run duration in milliseconds.
created_at	string | null	ISO 8601 timestamp when the run was created.
completed_at	string | null	ISO 8601 timestamp when the run reached terminal state.
metadata	object | null	The metadata object passed at run creation, if any.

• Retries: 3 attempts with exponential backoff (1s, 2s). Retries on 5xx / network errors only.
• Headers: Runflow-Request-Id is always sent. Runflow-Signature is sent if a signing secret is configured.

Additional Resources

• Playground
• API Reference
• Documentation
• OpenAPI

Related

Browse all models

Browse the catalog.

Run lifecycle

Callbacks, polling, statuses.

Callbacks

Handle async results.

Pricing

How requests bill out.