> ## Documentation Index
> Fetch the complete documentation index at: https://docs.slng.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# SLNG: the execution layer for real-time voice

> Speech to text, text to speech, and voice agents through one API. Use your existing models and orchestrator. Route each turn for lower cost and latency.

# SLNG documentation

Use regional speech models, the Context Router, and Agent Builder to build or improve real-time voice agents. Keep your existing orchestrator, prompts, tools, and provider relationships.

## Start here

* [Set up your account and API key](/guides/get-started/quickstart)
* [Deploy an agent on SLNG](/guides/get-started/create-a-project/create-agent)
* [Improve an existing agent](/guides/get-started/create-a-project/improve-agent)
* [Use the APIs directly](/guides/get-started/create-a-project/api-only)
* [Install the SLNG skills](/integrations/skills/agent-skill)
* [Browse available models](/models/catalog/all-models)
* [Read the API reference](/api-reference/authentication)

## Core products

* **Speech to Text API:** Transcribe audio files or live streams.
* **Text to Speech API:** Generate audio over HTTP or a streaming WebSocket connection.
* **Context Router:** Route each turn to full LLM reasoning, local inference, or a cached response.
* **Agent Builder:** Configure Listen, Think, and Speak, then deploy your agent to phone or web.
* **Execution Layer:** Reduce latency and model cost across Listen, Think, and Speak.