Google

Route to Google by prefixing the model with google/. gemini-3.1-flash-tts-preview is the newest and highest quality; the default gemini-2.5-flash-preview-tts favours latency.

Models

Usage

curl -X POST https://api.speechbase.ai/v1/audio/speech \
  -H "Authorization: Bearer $SPEECHBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "inline",
    "model": "google/gemini-2.5-flash-preview-tts",
    "voice": "Kore",
    "text": "Hello from Speechbase!",
    "output": "mp3"
  }' --output hello.mp3

Built-in voices include Kore, Puck, Zephyr, Charon, Aoede, Fenrir, and others — see the Gemini TTS docs for the full list.

Output format

Gemini returns raw PCM natively. Set the output field ("wav", "mp3", or "pcm") to have the gateway transcode for you; omit it to pass the raw PCM through. See Output formats.

Provider options

Anything in providerOptions is forwarded to the Gemini API unchanged (for example temperature).