Synthome Docs
Providers

Hume

Setup and configuration for Hume

Hume

Hume provides emotionally expressive text-to-speech with natural language emotion control.

Setup

1. Get API Key

  1. Create an account at hume.ai
  2. Go to your API Keys
  3. Create a new API key

2. Configure Environment

# .env or .env.local
HUME_API_KEY=your_hume_key_here

Available Models

ModelID
Hume TTShume/tts
import { generateAudio, audioModel } from "@synthome/sdk";

generateAudio({
  model: audioModel("hume/tts", "hume"),
  text: "I'm so excited to share this news with you!",
  description: "Excited, energetic, celebratory",
});

Emotion Control

Unlike other TTS providers, Hume uses natural language descriptions for emotion:

// Calm narration
generateAudio({
  model: audioModel("hume/tts", "hume"),
  text: "Take a deep breath and relax.",
  description: "Calm, soothing, gentle",
});

// Energetic announcement
generateAudio({
  model: audioModel("hume/tts", "hume"),
  text: "We just launched our new product!",
  description: "Excited, enthusiastic, professional",
});

Per-Request API Key

Override the environment variable for specific requests:

generateAudio({
  model: audioModel("hume/tts", "hume"),
  text: "Hello world",
  apiKey: "different_hume_key",
});

Resources

How is this guide?