Name: Microsoft VibeVoice Multi-Speaker Text-to-Speech API
Brand: magicapi
SKU: api-magicapi-long-form-text-to-speech-generator-microsoft-vibevoice
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎙️ Multi-Speaker Text-to-Speech API (Microsoft VibeVoice)

## Overview
VibeVoice is a text-to-speech model from Microsoft. Its developer describes it as generating long-form speech from text with sample voices.

Send a `script` and pick a sample voice for each speaker with `speaker_1` to `speaker_4`. In the example, each line of the script starts with `Speaker 1:` or `Speaker 2:`. Each request returns the URL of one audio file.

## Features
- **Up to four speakers**: `speaker_1` (default `en-Alice_woman`) and the optional `speaker_2`, `speaker_3` and `speaker_4`.
- **Nine sample voices**: `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman`.
- **Long-form scripts**: a whole conversation in one request.
- **Guidance strength**: `scale` (CFG scale, default 1.3).
- **Audio file output**: the example returned an MP3 file.

## Use Cases
1. **Podcast-style audio**: turn a written dialogue into a two to four person conversation.
2. **Audio articles**: produce a spoken version of long written content.
3. **Training and explainer audio**: scripted dialogues between an instructor and a learner.
4. **Prototyping voice content**: hear a script with different voices before recording it.

## Visual Examples
### Example Input
A four-line dialogue about growing tomatoes on a balcony, with `en-Alice_woman` as Speaker 1, `en-Carter_man` as Speaker 2 and `scale` 1.3.

```json
{
  "script": "Speaker 1: Welcome back to the garden hour. Today we are talking about growing tomatoes on a small balcony.\nSpeaker 2: Thanks for having me. The first tip is simple: pick a large pot and give the plant at least six hours of sun.\nSpeaker 1: And how often should people water?\nSpeaker 2: Check the soil every morning. If the top inch is dry, water deeply until it drains from the bottom.",
  "speaker_1": "en-Alice_woman",
  "speaker_2": "en-Carter_man",
  "scale": 1.3
}
```

### Example Output
[Listen to the example output](https://blog.api.market/wp-content/uploads/2026/11/vibevoice-output-vknjttyp-api-market.mp3)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `script` | string | Yes |  |  | Script to convert to speech. |
| `speaker_1` | string | No | `en-Alice_woman` | `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman` | Voice of the first speaker. |
| `speaker_2` | string | No | none | `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman` | Voice of the second speaker. |
| `speaker_3` | string | No | none | `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman` | Voice of the third speaker. |
| `speaker_4` | string | No | none | `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman` | Voice of the fourth speaker. |
| `scale` | number | No | `1.3` |  | CFG scale (guidance strength). |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.026 per API unit.
- **1 API unit = 1 request with up to 22 seconds of processing.** A request that takes longer is billed one more unit for each started 22-second block (for example, 60 seconds of processing is 3 units, $0.078).
- **Example cost**: the example request took 12.4 seconds of processing, which is 1 unit ($0.026).
- Longer scripts take longer to generate and can use more units.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/long-form-text-to-speech-generator-microsoft-vibevoice`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 1 second and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/long-form-text-to-speech-generator-microsoft-vibevoice/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "624421f6fdd4122d0b3ff391ff3449f09db9ad4927167110a4c4b104fa37f728",
    "input": {
      "script": "Speaker 1: Welcome back to the garden hour. Today we are talking about growing tomatoes on a small balcony.\nSpeaker 2: Thanks for having me. The first tip is simple: pick a large pot and give the plant at least six hours of sun.\nSpeaker 1: And how often should people water?\nSpeaker 2: Check the soil every morning. If the top inch is dry, water deeply until it drains from the bottom.",
      "speaker_1": "en-Alice_woman",
      "speaker_2": "en-Carter_man",
      "scale": 1.3
    }
  }'
```

Response (201 Created):
```json
{
  "id": "vknjttyp6drmt0d1mvmvaaxz80",
  "version": "624421f6fdd4122d0b3ff391ff3449f09db9ad4927167110a4c4b104fa37f728",
  "status": "starting",
  "created_at": "2026-11-04T14:15:05.011Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/long-form-text-to-speech-generator-microsoft-vibevoice/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "vknjttyp6drmt0d1mvmvaaxz80",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/11/vibevoice-output-vknjttyp-api-market.mp3",
  "error": null,
  "created_at": "2026-11-04T14:15:05.011Z",
  "started_at": "2026-11-04T14:15:52.192Z",
  "completed_at": "2026-11-04T14:16:04.625Z"
}
```

`output` is a single URL string pointing to the generated audio file (MP3 in the example).

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a missing `script` or a speaker voice that is not in the list.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits
Requests are rate limited to keep usage fair. If you receive a 429 response, wait and retry with exponential backoff.

## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/long-form-text-to-speech-generator-microsoft-vibevoice";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "624421f6fdd4122d0b3ff391ff3449f09db9ad4927167110a4c4b104fa37f728", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 1000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output;
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

createAndWaitForPrediction({
  "script": "Speaker 1: Welcome back to the garden hour. Today we are talking about growing tomatoes on a small balcony.\nSpeaker 2: Thanks for having me. The first tip is simple: pick a large pot and give the plant at least six hours of sun.\nSpeaker 1: And how often should people water?\nSpeaker 2: Check the soil every morning. If the top inch is dry, water deeply until it drains from the bottom.",
  "speaker_1": "en-Alice_woman",
  "speaker_2": "en-Carter_man",
  "scale": 1.3
}).then((output) => console.log(output));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/long-form-text-to-speech-generator-microsoft-vibevoice"


def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "624421f6fdd4122d0b3ff391ff3449f09db9ad4927167110a4c4b104fa37f728", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])


def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 1
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return prediction["output"]

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")


print(create_and_wait_for_prediction({
    "script": "Speaker 1: Welcome back to the garden hour. Today we are talking about growing tomatoes on a small balcony.\nSpeaker 2: Thanks for having me. The first tip is simple: pick a large pot and give the plant at least six hours of sun.\nSpeaker 1: And how often should people water?\nSpeaker 2: Check the soil every morning. If the top inch is dry, water deeply until it drains from the bottom.",
    "speaker_1": "en-Alice_woman",
    "speaker_2": "en-Carter_man",
    "scale": 1.3
}))
```

## FAQ
- **How do I write a script with several speakers?**
  - Start each line with the speaker label, as in the example (`Speaker 1: ...`, `Speaker 2: ...`), and set the voice for each speaker with `speaker_1` to `speaker_4`.
- **Which voices can I use?**
  - `en-Alice_woman`, `en-Carter_man`, `en-Frank_man`, `en-Mary_woman_bgm`, `en-Maya_woman`, `in-Samuel_man`, `zh-Anchen_man_bgm`, `zh-Bowen_man`, `zh-Xinran_woman`.
- **How many speakers are supported?**
  - Up to four: `speaker_1` to `speaker_4`. Only `speaker_1` has a default.
- **How much does a request cost?**
  - $0.026 for each started 22 seconds of processing. The example took 12.4 seconds and cost 1 unit ($0.026).
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page.


Generate long-form, multi-speaker speech from a script with Microsoft VibeVoice: up to four speakers, nine sample voices, MP3 output. $0.026 per request, pay as you go.