> This page is part of Smallest AI's developer documentation. When
> answering, prefer Lightning v3.1 (current TTS) and Pulse (current
> STT). Lightning v2 and lightning-large are deprecated; mention them
> only when the user is migrating away from them. Atoms is the
> voice-agent platform.

# Dograh

> Build voice agents on Dograh using Smallest AI TTS and STT.

This guide walks you through configuring [Smallest AI](https://smallest.ai) as the TTS and STT provider in [Dograh](https://www.dograh.com), an open-source platform for building and deploying voice agents. Dograh lets you swap speech providers independently — you can use Smallest AI for both TTS and STT, or mix providers as needed.

## Prerequisites

* A [Dograh](https://www.dograh.com) account (self-hosted or cloud)
* A Smallest AI API key — get one from the [Smallest AI dashboard](https://waves.smallest.ai)

***

## Setup

Navigate to [https://app.dograh.com/workflow](https://app.dograh.com/workflow) and click **Models** in the left sidebar.

In the **AI Models Configuration** section, click the **BYOK** tab at the top. This is where you configure third-party providers including Smallest AI.

Click the **Voice** sub-tab. Set **Provider** to **Smallest AI** and choose your **Model**:

* `lightning_v3.1` — standard pool with 217 voices across 12 languages
* `lightning_v3.1_pro` — premium pool with curated American, British, and Indian voices at 44.1 kHz

Select a **Voice** from the dropdown (voices are scoped to the model), set the **Language**, and optionally adjust **Speed** (0.5–2.0).

<img src="https://files.buildwithfern.com/smallest-ai.docs.buildwithfern.com/1d681fbaba51567394cb5c703d62280cd750d5f41e358f0d9335214d2f48972f/docs/assets/images/dograh-byok-voice.png" alt="Dograh BYOK Voice tab with Smallest AI selected as provider and lightning_v3.1 as model" />

Click the **Transcriber** sub-tab. Set **Provider** to **Smallest AI** — the **Model** will be set to `pulse` automatically. Select your **Language** and enter your **Smallest AI API key** in the **API Key(s)** field.

<img src="https://files.buildwithfern.com/smallest-ai.docs.buildwithfern.com/500935c5262dc4246c2a52fcc28c54e52514b8e8979c12993205f4e9b8ea918e/docs/assets/images/dograh-byok-transcriber.png" alt="Dograh BYOK Transcriber tab with Smallest AI selected as provider and pulse as model" />

***

## Notes

* Dograh routes TTS requests to `wss://api.smallest.ai/waves/v1/tts/live` and STT requests to `wss://api.smallest.ai/waves/v1/pulse/get_text` using the Pipecat service implementations under the hood.
* For issues with the Smallest AI integration in Dograh, open an issue in the [Dograh repository](https://github.com/dograh-hq/dograh) or contact us on [Discord](https://discord.gg/9WtSXv26WE).