> Revox API docs for developers and AI agents. Index: https://www.getrevox.com/docs/api/llms.txt · Everything in one file: https://www.getrevox.com/docs/api/llms-full.txt · OpenAPI 3.1: https://www.getrevox.com/openapi.json · MCP server: https://www.getrevox.com/api/mcp-servers/generic/mcp
> Base URL https://www.getrevox.com/api · Auth: `Authorization: Bearer $REVOX_API_KEY`

# Hear the voice say a word or one of its versions

`POST https://www.getrevox.com/api/assistants/{id}/pronunciations/say`

Returns the assistant's voice saying `text` exactly as sent, without applying any of its pronunciations: the word as written is how callers hear it before one is set.

## Before you call

- Nothing is saved: send the word as written to hear whether it needs a pronunciation, or a pronunciation's `replacement` to hear the one in use.

## Request

```bash
curl -X POST "https://www.getrevox.com/api/assistants/3f6c2a1e-8b4d-4c7a-9e2f-5d1b7a9c0e34/pronunciations/say" \
  -H "Authorization: Bearer $REVOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "text": "Swile" }'
```

## Path parameters

- `id` (string, required): ID of the assistant, as returned by [`GET /assistants`](https://www.getrevox.com/docs/api/assistants/get-assistants.md).

## Body

JSON.

- `text` (string, required): What the voice says: the word as written (how callers hear it with no pronunciation), or a version of it (`replacement` of a pronunciation, or a suggested version's `text`). 1 to 500 characters. Example: `"Swile"`.

## Response 200

```json
{ "audio_base64": "audio base64" }
```

- `audio_base64` (string): The assistant's voice saying `text` exactly as sent, base64-encoded WAV.

## Errors

- `404` Not found in your organization: `{"error":"Not found"}`
- Every status is explained under Errors in https://www.getrevox.com/docs/api.md#errors.