> For the complete documentation index, see [llms.txt](https://docs.delphi.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.delphi.ai/api-immortal-only/voice.md).

# Voice

Stream voice responses from your clone as raw PCM audio.

<a href="https://help.delphi.co" class="button primary">Visit the Help Center</a>

### Stream Voice Response

Send a text message to your clone and receive the response as a real-time audio stream.

**Endpoint:** `POST /v3/voice/stream`

**Prerequisites:**

* The clone must have a voice configured.
* A conversation must already exist (create one with `POST /v3/conversation` first).

**Request body:**

| Field             | Type   | Required | Description                          |
| ----------------- | ------ | -------- | ------------------------------------ |
| `conversation_id` | string | Yes      | UUID of an existing conversation     |
| `message`         | string | Yes      | User message (1 - 10,000 characters) |

**Example request:**

```bash
curl -X POST "https://api.delphi.ai/v3/voice/stream" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "conversation_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "message": "What are your thoughts on AI safety?"
  }' \
  --output response.pcm
```

**Response:**

Binary stream of raw PCM audio data (`application/octet-stream`).

**Response headers:**

| Header                    | Value               | Description             |
| ------------------------- | ------------------- | ----------------------- |
| `X-Audio-Format`          | `pcm_24000_16_mono` | Audio format identifier |
| `X-Audio-Sample-Rate`     | `24000`             | Sample rate in Hz       |
| `X-Audio-Bits-Per-Sample` | `16`                | Bit depth               |
| `X-Audio-Channels`        | `1`                 | Mono audio              |

**Playing the audio:**

```bash
# Convert PCM to WAV using ffmpeg
ffmpeg -f s16le -ar 24000 -ac 1 -i response.pcm response.wav
```

### Synthesize Voice

Convert raw text to audio using your clone's configured voice. Unlike Voice Stream, this does not\
generate a clone response — it speaks the exact text you provide.

**Endpoint:** `POST /v3/voice/synthesize`

**Prerequisites:**

* The clone must have a voice configured.

#### Request body

| Parameter | Type   | Required | Description                              |
| --------- | ------ | -------- | ---------------------------------------- |
| `text`    | string | Yes      | Text to synthesize (1–10,000 characters) |

#### Query parameters

| Parameter | Type    | Required | Default | Description                          |
| --------- | ------- | -------- | ------- | ------------------------------------ |
| `stream`  | boolean | No       | `false` | Stream raw PCM bytes instead of JSON |

#### Example request (batch)

```bash
curl -X POST "https://api.delphi.ai/v3/voice/synthesize" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello, this is a test of the synthesis endpoint."
  }'
```

#### Example response

```json
{
  "audio": "8/8AABcAHAAhAC8AKQAmADkALAA2AEgANwArACI..."
}
```

The `audio` field is base64-encoded raw PCM data (24 kHz, 16-bit signed, mono, little-endian).

#### Decoding the audio

```bash
echo "<base64_audio>" | base64 -d > output.pcm
ffmpeg -f s16le -ar 24000 -ac 1 -i output.pcm output.wav
```

#### Example request (streaming)

```bash
curl -X POST "https://api.delphi.ai/v3/voice/synthesize?stream=true" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello, this is a test of the streaming synthesis endpoint."
  }' \
  --output synthesized.pcm
```

#### Response

Binary stream of raw PCM audio data (`application/octet-stream`).

#### Response headers

| Header                    | Value               | Description             |
| ------------------------- | ------------------- | ----------------------- |
| `X-Audio-Format`          | `pcm_24000_16_mono` | Audio format identifier |
| `X-Audio-Sample-Rate`     | `24000`             | Sample rate in Hz       |
| `X-Audio-Bits-Per-Sample` | `16`                | Bit depth               |
| `X-Audio-Channels`        | `1`                 | Mono audio              |

#### Playing the audio

```bash
ffmpeg -f s16le -ar 24000 -ac 1 -i synthesized.pcm synthesized.wav
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.delphi.ai/api-immortal-only/voice.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.