Gradium

Gradium is a bundled text-to-speech provider for OpenClaw. It can generate normal audio replies, voice-note-compatible Opus output, and 8 kHz u-law audio for telephony surfaces.

Setup

Create a Gradium API key, then expose it to OpenClaw:

bash
Copy
export GRADIUM_API_KEY="gsk_..."

You can also store the key in config under

.

Config

json5
Copy
{
  messages: {
    tts: {
      auto: "always",
      provider: "gradium",
      providers: {
        gradium: {
          voiceId: "YTpq7expH9539ERJ",
          // apiKey: "${GRADIUM_API_KEY}",
          // baseUrl: "https://api.gradium.ai",
        },
      },
    },
  },
}

Voices

Default voice: Emma.

Output

• Audio-file replies use WAV.
• Voice-note replies use Opus and are marked voice-compatible.
• Telephony synthesis uses
  text

  Copy

  ulaw_8000
  at 8 kHz.

Related

• Text-to-Speech
• Media Overview