> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ortavox.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ortavox.ai/_mcp/server.

# Transcriber

> Configure the Speech-to-Text engine for your Ortavox Agent

The `transcriber` object defines the Speech-to-Text (STT) engine that converts spoken audio into text in real time. It is a **required** field when creating an Agent.

## Schema

```json
{
  "transcriber": {
    "provider": "deepgram",
    "model": "nova-3",
    "language": "en"
  }
}
```

## Parameters

| Property   | Type     | Required | Default | Description                                  |
| ---------- | -------- | -------- | ------- | -------------------------------------------- |
| `provider` | `string` | ✅        | —       | STT provider (see supported providers below) |
| `model`    | `string` | ✅        | —       | Provider-specific model name (e.g. `nova-3`) |
| `language` | `string` | ✅        | `en`    | BCP-47 language code (e.g. `en`, `fr`, `es`) |

## Supported Providers

| Provider     | `provider` value | Notes                                             |
| ------------ | ---------------- | ------------------------------------------------- |
| Deepgram     | `deepgram`       | Recommended — ultra-low latency, highest accuracy |
| OpenAI       | `openai`         | Whisper-powered                                   |
| AssemblyAI   | `assemblyai`     | Strong for noisy environments                     |
| Google       | `google`         | Multi-language support                            |
| Speechmatics | `speechmatics`   | Specialised across 50+ languages                  |
| Azure        | `azure`          | Enterprise-grade                                  |