数字人、AI 口播、异兽、舞蹈与绘画：多模态内容生产工程化实践

import os, time, requests

API_BASE = os.environ["AIGC_API_BASE"]
HEADERS = {
    "Authorization": f"Bearer {os.environ['AIGC_API_KEY']}",
    "Content-Type": "application/json",
}

def submit(path: str, payload: dict) -> str:
    r = requests.post(f"{API_BASE}{path}", json=payload,
                      headers=HEADERS, timeout=60)
    r.raise_for_status()
    return r.json()["task_id"]

def poll(path: str, task_id: str, timeout=300):
    start = time.time()
    while time.time() - start < timeout:
        r = requests.get(f"{API_BASE}{path}/{task_id}",
                         headers=HEADERS, timeout=30)
        data = r.json()
        if data["status"] == "succeeded":
            return data
        if data["status"] == "failed":
            raise RuntimeError(data.get("error"))
        time.sleep(4)
    raise TimeoutError(task_id)

def image_prompt(scene: str, character: str, style: str) -> str:
    return (
        f"{style}，{character}，{scene}，"
        "电影级光影，细节清晰，构图干净，"
        "无文字水印，无多余肢体，无恐怖元素"
    )

def gen_image(scene, character, style="国风插画，赛璐璐上色"):
    prompt = image_prompt(scene, character, style)
    task_id = submit("/v1/images/tasks", {
        "prompt": prompt, "ratio": "9:16"
    })
    return poll("/v1/images/tasks", task_id)["image_url"]

def tts(text: str, voice="female_01") -> str:
    task_id = submit("/v1/tts/tasks", {
        "text": text, "voice": voice
    })
    return poll("/v1/tts/tasks", task_id)["audio_url"]

def digital_human(avatar_id: str, audio_url: str) -> str:
    task_id = submit("/v1/avatar/tasks", {
        "avatar_id": avatar_id,
        "audio_url": audio_url,
        "background": "studio",
    })
    return poll("/v1/avatar/tasks", task_id)["video_url"]

def gen_dance(ref_image: str, motion: str, duration=8) -> str:
    task_id = submit("/v1/video/tasks", {
        "image_url": ref_image,
        "motion": motion,
        "duration": duration,
        "ratio": "9:16",
    })
    return poll("/v1/video/tasks", task_id)["video_url"]

# 视频 + 音频 + 字幕
ffmpeg -i video.mp4 -i voice.mp3 -vf "subtitles=sub.srt" \
  -c:v libx264 -c:a aac -shortest final.mp4