joshu

Google: Gemini 3.8 Flash TTS

From Google AI (Gemini API), through Joshu Data Bodega.

Part of Voice generation gemini 3 8 flash tts generate.

Price
$0.055 per call

No calls measured in the last 30 days.

Prices include the router's 10% markup and are paid from the usage wallet. Checked on November 11, 2026.

Inputs

  • contentsarray, required

    One user turn: {parts: [{text: script}]}. Gemini 3.8 TTS reads `text` as a verbatim transcript: a direction written into it ("Say cheerfully: ...") is spoken aloud and billed as audio. Put delivery directions in the part's `speechMetadata.style` ({text: "Welcome to treg.", speechMetadata: {style: cheerful}}) and keep `text` to the words to be spoken. For two speakers, give each line its own part with its speaker: [{text: "Did it finish?", speechMetadata: {speaker: Joe}}, {text: "Yes.", speechMet

  • generationConfigobject, required

Call it

Router call with a router API key
curl https://hello.joshu.me/api/router/call \
  -H "Authorization: Bearer $JOSHU_ROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "endpoint": "google-ai.voice-gen.gemini-3-8-flash-tts",
    "params": {
      "contents": [
        {
          "parts": [
            {
              "text": "Welcome to treg, the tool catalog for your agents.",
              "speechMetadata": {
                "style": "cheerful"
              }
            }
          ]
        }
      ],
      "generationConfig": {
        "speechConfig": {
          "voiceConfig": {
            "prebuiltVoiceConfig": {
              "voiceName": "Kore"
            }
          }
        },
        "responseModalities": [
          "AUDIO"
        ]
      }
    }
  }'