Name: P-Video 2 Pro Avatar API: Talking Avatar Video Generator
Brand: magicapi
SKU: api-magicapi-talking-avatar-video-creator-prunaai-p-video-2-pro-avatar
Availability: InStock
Author: API Market

Shashank Agarwal

# 🗣️ Talking Avatar Video Generator API (Pruna AI P-Video 2 Pro Avatar)

## Overview
P-Video 2 Pro Avatar is a talking-avatar model from Pruna AI. Its developer describes it as turning a portrait and an audio file or script into a talking video with synced lips and expressions, from 1 second to 3 minutes long.

Send a portrait as `image` and either a speech recording as `audio` or the words to speak as `voice_script` (spoken by a built-in Gemini TTS voice). `prompt` describes what the video should show. Each request returns the URL of one MP4 video at 480p or 768p.

## Features
- **Two ways to add speech**: send your own recording as `audio` (up to 3 minutes; the video matches its length), or send `voice_script` and the model speaks it with a TTS voice. `audio` overrides `voice_script`.
- **30 TTS voices**: `voice` offers 30 named voices, female and male (default `Zephyr (Female)`), with delivery instructions in `voice_prompt` (tone, pacing, accent, emotion).
- **10 script languages**: `voice_language` is English (US) (default), English (UK), Spanish, French, German, Italian, Portuguese (Brazil), Japanese, Korean or Hindi.
- **Scene direction**: `prompt` describes what the video shows (default "The person is talking.").
- **Prompt upsampling**: with `prompt_upsampling` on (default), a vision model looks at the image and appends the scene's own motion (rain, lights, background, body movement) to your prompt. Your words stay unchanged and come first.
- **Resolution**: `480p` or `768p` (default).

## Use Cases
1. **Spokesperson videos**: turn a portrait and a voice-over into a presenter clip.
2. **Localized messages**: speak the same script in one of ten languages with `voice_script` and `voice_language`.
3. **Character and social content**: animate a portrait to a recorded line with a scene described in `prompt`, as in the example.
4. **Course and help videos**: narrate short lessons with a talking presenter of up to 3 minutes.

## Visual Examples
### Example Input
![Input portrait](https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-2-api-market.webp)

[Listen to the input audio](https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-api-market.mp3)

Prompt: "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room's soft, warm lighting stays the same throughout. One continuous shot, no cuts."

`resolution` is `768p`, `seed` is `42` and `prompt_upsampling` is `false`. Because `audio` is set, `voice_script` is empty and the TTS settings (`voice`, `voice_prompt`, `voice_language`) are not used for speech.

### Example Output
An 8-second 768p talking video of the portrait speaking the input audio.

[Watch the example output](https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-output-sanwjcwh-api-market.mp4)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `image` | string (URL) | Yes | none |  | Portrait of the person who speaks, used as the first frame (jpg, jpeg, png or webp URL) |
| `audio` | string (URL) | No | none |  | Speech audio URL, up to 3 minutes; the video matches its length. Overrides `voice_script` |
| `voice_script` | string | No | `""` |  | Exact words to speak with Gemini TTS. Required when no `audio` is sent |
| `voice` | string | No | `Zephyr (Female)` | `Zephyr (Female)`, `Puck (Male)`, `Charon (Male)`, `Kore (Female)`, `Fenrir (Male)`, `Leda (Female)`, `Orus (Male)`, `Aoede (Female)`, `Callirrhoe (Female)`, `Autonoe (Female)`, `Enceladus (Male)`, `Iapetus (Male)`, `Umbriel (Male)`, `Algenib (Male)`, `Despina (Female)`, `Erinome (Female)`, `Laomedeia (Female)`, `Achernar (Female)`, `Algieba (Male)`, `Schedar (Male)`, `Gacrux (Female)`, `Pulcherrima (Female)`, `Achird (Male)`, `Zubenelgenubi (Male)`, `Vindemiatrix (Female)`, `Sadachbia (Male)`, `Sadaltager (Male)`, `Sulafat (Female)`, `Alnilam (Male)`, `Rasalgethi (Male)` | Gemini TTS voice for `voice_script` |
| `voice_prompt` | string | No | `Say the following.` |  | Delivery instructions for the TTS voice (tone, pacing, accent, emotion). Not spoken |
| `voice_language` | string | No | `English (US)` | `English (US)`, `English (UK)`, `Spanish`, `French`, `German`, `Italian`, `Portuguese (Brazil)`, `Japanese`, `Korean`, `Hindi` | Language of `voice_script` |
| `prompt` | string | No | `The person is talking.` |  | What the video should show |
| `resolution` | string | No | `768p` | `480p`, `768p` | Video resolution |
| `seed` | integer | No | none |  | Random seed for reproducible generation |
| `prompt_upsampling` | boolean | No | `true` |  | Let a vision model look at the image and append the scene's own motion (rain, lights, background, body movement) to your prompt. Your words stay unchanged and come first |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.025 per API unit.
- **1 API unit = 1 second of output video at 480p**, rounded up per request. A 768p second counts as 2 units.
- The video is as long as the speech (your `audio`, or the spoken `voice_script`), from 1 second to 3 minutes.
- Price per second of video by `resolution`:

  | `resolution` | Factor | Price per second |
  | --- | --- | --- |
  | `480p` | 1 | $0.025 |
  | `768p` (default) | 2 | $0.05 |
- A 10-second clip costs $0.25 at 480p and $0.50 at 768p; a 3-minute clip costs $4.50 at 480p and $9 at 768p.
- **Example cost**: the example produced 8 seconds of 768p video: 8 x $0.05 = **$0.40**.
- **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/talking-avatar-video-creator-prunaai-p-video-2-pro-avatar`

### 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 2 seconds and an overall timeout of 900 seconds (15 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/talking-avatar-video-creator-prunaai-p-video-2-pro-avatar/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "prunaai/p-video-2-pro-avatar",
    "input": {
      "seed": 42,
      "audio": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-api-market.mp3",
      "image": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-2-api-market.webp",
      "voice": "Zephyr (Female)",
      "prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room'\''s soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
      "resolution": "768p",
      "voice_prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room'\''s soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
      "voice_script": "",
      "voice_language": "English (US)",
      "prompt_upsampling": false
    }
  }'
```

Response (201 Created):
```json
{
  "id": "sanwjcwhzhrmy0d1d18vsjy9yc",
  "version": "prunaai/p-video-2-pro-avatar",
  "status": "starting",
  "created_at": "2026-10-23T10:32:18.855Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/talking-avatar-video-creator-prunaai-p-video-2-pro-avatar/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "sanwjcwhzhrmy0d1d18vsjy9yc",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-output-sanwjcwh-api-market.mp4",
  "error": null,
  "created_at": "2026-10-23T10:32:18.855Z",
  "started_at": "2026-10-23T10:32:19.355Z",
  "completed_at": "2026-10-23T10:32:40.997Z"
}
```

`output` is a single URL string pointing to the generated MP4 video.

### 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 `image`, no `audio` and an empty `voice_script`, or a `voice` or `resolution` 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
All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## 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/talking-avatar-video-creator-prunaai-p-video-2-pro-avatar";

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: "prunaai/p-video-2-pro-avatar", 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 = 900000;
  const pollInterval = 2000;
  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 15 minutes");
}

createAndWaitForPrediction({
  "seed": 42,
  "audio": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-api-market.mp3",
  "image": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-2-api-market.webp",
  "voice": "Zephyr (Female)",
  "prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room's soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
  "resolution": "768p",
  "voice_prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room's soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
  "voice_script": "",
  "voice_language": "English (US)",
  "prompt_upsampling": false
}).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/talking-avatar-video-creator-prunaai-p-video-2-pro-avatar"


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": "prunaai/p-video-2-pro-avatar", "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 = 900
    poll_interval = 2
    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 15 minutes")


output = create_and_wait_for_prediction({
    "seed": 42,
    "audio": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-api-market.mp3",
    "image": "https://blog.api.market/wp-content/uploads/2026/10/p-video-2-pro-avatar-input-2-api-market.webp",
    "voice": "Zephyr (Female)",
    "prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room's soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
    "resolution": "768p",
    "voice_prompt": "The woman whispers softly and intimately to the camera, like in an ASMR video, with a warm, playful smile. She leans in slightly toward the camera and tilts her head a little as she speaks. The room's soft, warm lighting stays the same throughout. One continuous shot, no cuts.",
    "voice_script": "",
    "voice_language": "English (US)",
    "prompt_upsampling": False
})
print(output)
```

## FAQ
- **Do I need my own audio?**
  - No. Leave `audio` empty and send the words in `voice_script`; a TTS voice (`voice`) speaks them in `voice_language`. If you send `audio`, it overrides `voice_script`.
- **How long can the video be?**
  - From 1 second to 3 minutes. The video matches the length of the speech, and you are billed per second of output video.
- **Is `voice_prompt` spoken?**
  - No. It only tells the TTS voice how to deliver `voice_script` (tone, pacing, accent, emotion).
- **What does `prompt_upsampling` do?**
  - When on (the default), a vision model adds the motion it sees in your image (rain, lights, background, body movement) after your prompt. Turn it off to use your prompt alone.
- **How much does 768p cost compared with 480p?**
  - 768p costs $0.05 per second and 480p $0.025 per second. 768p is the default.
- **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. You can also email our support team at hello@magicapi.com.


Create lip-synced talking avatar videos from a portrait and your audio or a TTS script with Pruna AI's P-Video 2 Pro Avatar. Pay as you go, no monthly fee.