Basics

Getting Started with slng.ai

This guide will walk you through your first voice API call using slng.ai. Whether you're building a voicebot, adding accessibility features, or running audio workflows at scale, we make it simple to get started.


🧾 1. Get Your API Key

Sign up at slng.ai and retrieve your API key from the dashboard.

🔐 Keep your API key secret. It authenticates your account and usage.


🧪 2. Make Your First API Call

You can test the API right away using curl, Postman, or any HTTP client.

Here's a simple example using the VUI model for text-to-speech:

Code(bash)
curl -X POST https://api.slng.ai/tts/vui \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "text": "Hello world, from slng dot ai" }'

This will return an audio file (.mp3 or .wav) as a binary stream or base64, depending on your configuration.


🧠 3. Or Let slng.ai Pick the Best Model

Instead of choosing a model yourself, just call the base endpoint:

Code(bash)
curl -X POST https://api.slng.ai/tts \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "text": "Auto-routed voice response", "language": "en-US", "region": "eu-west" }'

We'll smart-route the request to the most suitable model based on:

  • Your language and region
  • Quality preferences
  • Pricing limits
  • Voice cloning options
  • Account preferences

🎚️ 4. Adjust Request Parameters

You can customize output via optional parameters:

Code(json)
{ "text": "My custom voice example", "language": "en-GB", "region": "eu-west", "clone": true, "pricing_limit": 0.01, "quality_minimum": 0.85 }

⚙️ 5. Try Speech-to-Text (STT)

To convert audio into text, send a request like this:

Code(bash)
curl -X POST https://api.slng.ai/stt/whisper-v3 \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: multipart/form-data" \ -F "[email protected]"

Or use the smart-routed /stt endpoint with optional parameters.


🔁 6. What’s Next?


📞 Help & Support

Need help? We're here:


Let your app speak, transcribe, and listen — with just a few lines of code.

Last updated: June 2025

Last modified on