Gradium

OverviewCopy Anchor Link

• GradiumTTSService: streams text over a WebSocket to Gradium and pushes 48 kHz audio as it arrives.

InstallationCopy Anchor Link

bun add @pipedog/gradium

import { GradiumTTSService } from "@pipedog/gradium/tts";

PrerequisitesCopy Anchor Link

• A Gradium account and an API key from Gradium.
• Pass the key as apiKey. The examples read it from GRADIUM_API_KEY.

GradiumTTSServiceCopy Anchor Link

Connects to wss://api.gradium.ai/api/speech/tts, with the key in the x-api-key header.

ConfigurationCopy Anchor Link

SettingsCopy Anchor Link

Runtime-updatable settings, passed as settings: new GradiumTTSService.Settings({...}) (GradiumTTSSettings). Any setting you leave out keeps the default below.

UsageCopy Anchor Link

import { GradiumTTSService } from "@pipedog/gradium/tts";

const tts = new GradiumTTSService({
  apiKey: process.env.GRADIUM_API_KEY ?? "",
  settings: new GradiumTTSService.Settings({ voice: "_6Aslh2DxfmnRLmP" }),
});

NotesCopy Anchor Link

• The sample rate is fixed at 48000 Hz.
• jsonConfig takes a JSON string of extra model settings, sent with the setup message.

Generated from pipecat.runtimes and pipecat.compat in packages/gradium/package.json: ✅ means the compat smoke test passes on that runtime.

Runtime notesCopy Anchor Link

• The WebSocket goes through the core websocketConnect, with headers on the handshake.
• The deprecated InputParams, params, voice_id and model constructor options are not ported.
• Tested against a local WebSocket server (upstream has no Gradium TTS test), not against Gradium itself.