Pruna AI

P-Video 2 Pro Avatar

p-video-2-pro-avatar

Pruna AI's portrait-driven talking video model generates lip-synced, expressive talking-head clips from a single portrait image. It can be driven by uploaded audio or read a script aloud directly using built-in text-to-speech, making it well suited for batch production of course explainers, multilingual dubbing, and digital human short videos

p-video-2-pro-avatar

Try p-video-2-pro-avatar

Input

Showcases

View all showcases

Output

video

Fill in the parameters and run — results show up here.

Pricing

p-video-2-pro-avatar per second

Specifications	Unit Price
480P	$0.0286
768P	$0.0572
1080P	$0.0858
2K	$0.1143
4K	$0.1715

Positioning Overview

A video generation model for digital human talking-head scenarios: upload a single portrait as the first frame, then provide speech audio or input script text to generate a video of the person speaking. The script is read aloud by built-in speech synthesis, with multiple voice options and languages, and prompts describe the person's movements and expressions while speaking. It is suitable for content creators who need a real-person on-camera presence, such as course explanations, marketing talking-head videos, and short social videos, eliminating the need for filming and lip-syncing.

Core Highlights

A single portrait drives a talking video — only one portrait is needed as the first frame, with no existing video footage required, to generate footage of the person speaking.
Two driving modes: audio and script — you can upload speech audio of no more than 3 minutes, or input a script that is read aloud by built-in speech synthesis, with the video length following the speech.
Prompts control expressions and movements — use prompts to describe the person's appearance, movements, and expressions while speaking, and set a random seed to reproduce results.
Multiple resolution tiers cover production needs — choose from five resolution tiers: 480p, 768p, 1080p, 2k, and 4k, making it easy to select image quality based on the publishing platform.

Capability Description

Portrait-driven talking video: provide a portrait in jpg, jpeg, png, or webp format via an HTTPS link as the first frame to generate a video of the person speaking, with the visual content described by a prompt covering their appearance, movements, and expressions.

Upload audio to drive lip-sync: provide an HTTPS speech audio link; the video length matches the audio, up to 3 minutes; when using the 4k specification, the audio can be up to 125 seconds. In this case, the script, voice, language, and speech style settings no longer take effect.

Script-to-speech driving: when no audio is provided, input a script of up to 5000 characters, which is read aloud by built-in speech synthesis to generate the video; the script language must be selected from ten given languages, and a speech style description of up to 1000 characters can be used to describe the tone and pacing.

Voice and visual parameter selection: choose one of 30 built-in voices to read the script, use a prompt of up to 2000 characters to describe the visuals, and set a random seed between 0 and 2147483647; choose one resolution from 480p, 768p, 1080p, 2k, and 4k, with 768p as the default.

Applicable Scenarios

Batch talking-head videos for course instructors — instructors provide a portrait and explanation script to batch-generate on-camera explanation videos for online courses or knowledge programs.
Real-person talking-head marketing assets — operations teams use a product spokesperson's portrait together with promotional copy to generate talking-head clips for feed ads or brand social accounts.
Turning existing recordings into finished videos — creators upload their own recordings and portraits so the on-screen lip-sync matches the recording, directly producing short videos for social platforms.
Converting static portraits into talking videos — use existing photos of people together with scripts or audio to generate speaking footage with expressions and lip-sync for digital human programs or showcase pages.

Limitations and Selection Advice

Scripts and prompts have length limits: prompts up to 2000 characters, scripts up to 5000 characters, and speech style descriptions up to 1000 characters. The built-in voices do not support Chinese; the script language must be selected from English, Spanish, French, German, Italian, Portuguese (Brazil), Japanese, Korean, and Hindi. Video length is determined by the speech, up to 3 minutes; when using the 4k specification, the audio can be up to 125 seconds.

What's new

View changelog →

New model P-Video 2 Pro Avatar is live: turn one portrait plus audio or a script into a talking video up to 3 minutes long Nov 12, 2026

Frequently asked questions

View all FAQs →

What audio can be used to drive the digital human?

Pass an HTTPS audio link via audio_url. The video length equals the audio duration, up to a maximum of 3 minutes (4k up to a maximum of 125 seconds). Once audio is provided, script, voice, voice_language, and voice_prompt no longer take effect; if both are provided, the audio takes precedence.
How long can the script text be?

The script field allows up to 5000 characters, but the video length is ultimately determined by the duration of the built-in TTS reading the script aloud, with an upper limit of 3 minutes; when approaching the limit, it is recommended to first trim the text or generate it in segments.
How can I write better prompts for talking-head videos?

The prompt describes the person's appearance, movements, and demeanor while speaking, without writing any dialogue. It is recommended to write in the order of "person's appearance → range of movement → expression and gaze → camera feel", for example, "a man in a dark suit, sitting naturally, occasionally nodding, smiling at the camera, fixed medium shot", kept within 2000 characters; the default value is only "The person is talking."
Which languages and voices are supported?

voice_language supports 10 languages: American English, British English, Spanish, French, German, Italian, Brazilian Portuguese, Japanese, Korean, and Hindi; Chinese is not currently included. voice offers 30 preset voices, with gender indicated in the name; the default is Zephyr (Female). The text in script must match the selected language.

Create a generation task

POST https://api.hiapi.ai/v1/tasks

Headers

AuthorizationstringRequired

Account API key. Required on every request. Example: Bearer YOUR_HIAPI_API_KEY. Get your API key

Content-TypestringOptional

The request body is always parsed as JSON; when sent, only application/json (optionally with charset) is accepted — any other value returns 415. Example: application/json.

Idempotency-KeystringOptional

Idempotency key, up to 255 bytes. Retries with the same key from the same account create the task only once; replays return the original taskId.

Idempotency key

On network timeouts or client retries, the same key creates the task only once — no double charges (scope: the API key's account). Use a business-unique identifier as the key (e.g. "order-42"); replays return the original taskId, and keys expire 24 hours after the first request. See the Unified Async API overview for the full scenario matrix.

Body

modelstringRequired

Current fixed value: p-video-2-pro-avatar

routestringOptional

Current fixed value: default

inputobjectRequired

Business parameters for p-video-2-pro-avatar and route default — see the parameter table below.

seedintegerOptionalmin 0max 2147483647

Random seed, an integer from 0–2147483647; passing the same seed while keeping other inputs unchanged helps reproduce results. If not provided, it is random.

voicestringOptionalDefault: "Zephyr (Female)"

The voice used for the built-in text-to-speech, default Zephyr (Female). This parameter is ignored when audio_url is provided.

Options:Zephyr (Female)Puck (Male)Charon (Male)Kore (Female)Fenrir (Male)Leda (Female)Orus (Male)Aoede (Female)

Show all (22)Show less

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)

promptstringOptionalDefault: "The person is talking."maxLength 2000

A description of the video content, explaining the person's appearance, actions, and behavior while speaking; default is "The person is talking.".

scriptstringOptionalmaxLength 5000

The exact script the digital human will read aloud, written in the selected voice_language (Chinese is not supported). Required when audio_url is not provided; the reading duration determines the video length, up to 3 minutes.

audio_urlstringOptionalpattern ^https://

An HTTPS URL of the speech audio, up to 3 minutes (up to 125 seconds at 4k resolution); the video length matches the audio. When provided, script, voice, voice_language, and voice_prompt are all ignored.

image_urlstringRequiredpattern ^https://

An HTTPS URL of the speaking person's portrait image, used as the video's first frame; supports jpg, jpeg, png, and webp formats.

resolutionstringOptionalDefault: "768p"

Output resolution, options are 480p, 768p, 1080p, 2k, 4k, default 768p; higher resolution costs more.

Options:480p768p1080p2k4k

voice_promptstringOptionalDefault: "Say the following."maxLength 1000

A voice style instruction describing the delivery such as tone, pacing, accent, or emotion; it is not read aloud. Default is "Say the following.". Ignored when audio_url is provided.

voice_languagestringOptionalDefault: "English (US)"

The language used for the script text, default English (US). Ignored when audio_url is provided.

Options:English (US)English (UK)SpanishFrenchGermanItalianPortuguese (Brazil)Japanese

Show all (2)Show less

KoreanHindi

callbackobjectOptional

Terminal-state notification config. When omitted, no callback is delivered — poll the task yourself. See the callback note below.

Callbacks (Webhook)

Provide callback.url when creating a task and HiAPI POSTs to your service once the task reaches a terminal state — no polling needed. The callback payload mirrors the task detail structure; signature verification, delivery retries and idempotency are documented on the Unified Async API overview page.

urlstringRequired

Required when callback is present. A POST is sent to this URL once the task reaches a terminal state; must be a valid http(s) URL.

whenstringOptionalDefault: "final"

Only final is supported today — notified on both success and fail.

Options:final

storagestringOptionalDefault: "temp"

Artifact storage tier: temp (default, retained about 7 days) or persistent (long-term, billed by size). Insufficient balance or exceeding the storage cap silently downgrades to temp; the effective tier is shown in the task detail response.

Options:temppersistent

Request examples

# Submit an async task; returns a taskId immediately (no waiting for generation)
curl -X POST "https://api.hiapi.ai/v1/tasks" \
  -H "Authorization: Bearer $HIAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "p-video-2-pro-avatar",
  "input": {
    "voice": "Zephyr (Female)",
    "prompt": "The person is talking.",
    "image_url": "…",
    "resolution": "768p",
    "voice_prompt": "Say the following.",
    "voice_language": "English (US)"
  }
}'

const res = await fetch("https://api.hiapi.ai/v1/tasks", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.HIAPI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "p-video-2-pro-avatar",
    input: {
      "voice": "Zephyr (Female)",
      "prompt": "The person is talking.",
      "image_url": "…",
      "resolution": "768p",
      "voice_prompt": "Say the following.",
      "voice_language": "English (US)"
    }
  }),
});
const task = await res.json();
// { code: 200, message: "success", data: { taskId } }

import os, requests

res = requests.post(
    "https://api.hiapi.ai/v1/tasks",
    headers={"Authorization": f"Bearer {os.environ['HIAPI_API_KEY']}"},
    json={
        "model": "p-video-2-pro-avatar",
        "input": {
          "voice": "Zephyr (Female)",
          "prompt": "The person is talking.",
          "image_url": "…",
          "resolution": "768p",
          "voice_prompt": "Say the following.",
          "voice_language": "English (US)"
        }
    },
)
print(res.json())  # { code, message, data: { taskId } }

200 OK

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "tk-hiapi-01HZTQ8BX2N3GM3YFK4Z9D7VQR"
  }
}

400 Bad Request

{
  "code": 400,
  "message": "invalid request",
  "data": null,
  "error_code": "INVALID_REQUEST"
}

Get task detail

GET https://api.hiapi.ai/v1/tasks/:id

Response fields

taskIdstringRequired

Task ID, always prefixed with tk-hiapi-.

modelstringRequired

Resolved full model name. Tasks submitted with the route parameter return base@route (e.g. gpt-image-2/text-to-image@ext); otherwise identical to the name passed at creation.

routestringOptional

Present only when the task was submitted with an explicit route parameter; echoes the submitted value.

statusstringRequired

Task status.

Status values

queued Queued, not started yet. handling Being processed. archiving Processing finished, artifacts being prepared. success Terminal: completed, artifacts available (output returned). fail Terminal: failed (error returned). After submission, check status: success and fail are terminal, everything else is non-terminal. Poll every 1–3 seconds until terminal. On success, download artifacts from output[].url and transfer them before expireAt; pass storage: "persistent" at creation time when long-term retention is needed.

creatednumberRequired

Task creation time (Unix seconds).

completednumberRequired

Unix seconds when the task reached a terminal state; fixed to 0 while not terminal.

outputarrayOptional

Present only when status == success. Artifact list — always an array; a single task may return multiple types at once (e.g. video + first_frame).

Array [ object ]

Array items

urlstringRequired

Artifact URL, directly downloadable.

typestringRequired

Current possible values: video, image, audio, first_frame, last_frame, thumbnail; more may be added later.

expireAtnumberRequired

Unix seconds until which the artifact remains accessible; download or transfer before it expires.

errorobjectOptional

Present only when status == fail.

codestringRequired

Business error code.

messagestringRequired

Error description.

Response examples

Success

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "tk-hiapi-01HZTQ8BX2N3GM3YFK4Z9D7VQR",
    "model": "happyhorse-1-0",
    "status": "success",
    "created": 1777282033,
    "completed": 1777282099,
    "output": [
      {
        "url": "https://cdn.hiapi.ai/tasks/.../video.mp4",
        "type": "video",
        "expireAt": 1777886899
      }
    ]
  }
}

Processing (non-terminal)

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "tk-hiapi-...",
    "model": "happyhorse-1-0",
    "status": "archiving",
    "created": 1777282033,
    "completed": 0
  }
}

Failed

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "tk-hiapi-...",
    "model": "happyhorse-1-0",
    "status": "fail",
    "created": 1777282033,
    "completed": 1777282099,
    "error": {
      "code": "TASK_FAILED",
      "message": "task failed"
    }
  }
}