SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsbyted-mediakit

byted-mediakit

Design & creation

View repository

Updated 10-15 17:30

About this skill

Volcengine AI MediaKit

Prerequisites

  • Python: Confirm python --version ≥ 3.6
  • Dependencies: Script dependencies python-dotenv

Parameter Passing Methods

All scripts support two JSON parameter passing methods:

  1. Inline JSON(suitable for simple parameters):python script.py '{"key":"value"}'
  2. File Reference(recommended, avoids shell escaping issues):python script.py @params.json

@ The prefix indicates reading JSON content from a file; the file path is relative to the current working directory.

Result Delivery Rules

  • After successfully submitting an asynchronous task, an asynchronous task id will be returned, with the field being VCCreativeId or TaskId. When delivering the final artifact to the user,you mustinclude the asynchronous task id
  • When displaying the final artifact link,it is prohibitedto arbitrarily modify the link content
  • Give priority toproviding the artifact link to the user

Workflow

1) Identify the input video type (if necessary, upload first to obtain vid://...）

All subsequent processing scriptsprioritize using VOD-side resource references：

  • Vid：vid://vxxxx(or some scripts accept a bare vxxxx and automatically prepend vid://）
  • DirectUrl / FileName：directurl://<vod_file_name>(for media-type tasks, when using DirectUrl , it will require FileName + SpaceName）

When the user provides one of the following inputs, you need to execute the upload logic first, obtain Vid , and then continue:

  • Local file path: such as /path/to/a.mp4
  • http/https link: such as https://example.com/a.mp4(will perform URL fetch upload and poll the upload result)

Uniformly use scripts/upload_media.py：

python <SKILL_DIR>/scripts/upload_media.py "<local_file_path_or_http_url>" [space_name]

In the script output, the Source field is vid://..., which can be directly used as input for subsequent processing.

Security restriction: Local file upload only allows files under the workspace/, userdata/, and /tmp directories.

2) Identify user intent → select the corresponding processing script

Based on user needs, select a script according to the following decision tree:

User intent Script
Combine multiple videos/audios into one (sequential concatenation) stitching
Extract a certain time segment from a video/audio clipping
Speed up/slow down/change speed speedup
Mirror/flip vertically/flip horizontally flip
Generate a video by connecting multiple images in sequence image_to_video
Replace/overlay the background music of a video compile
Only want the audio track from the video extract_audio
Overlay and play multiple audio tracks simultaneously (mixing) mix_audios
Separate vocals and accompaniment/background sound voice_separation
Remove environmental noise/electrical noise/wind noise noise_reduction
Blurry/low-quality video restoration (compression artifacts/noise/scratches) quality_enhance
Low-resolution video enhancement (such as 720P→1080P) super_resolution
Frame interpolation for low-frame-rate videos to improve smoothness (such as 30fps→60fps) interlacing
Speech recognition/ASR/extract text dialogue from video asr_speech_to_text
OCR text extraction/recognize on-screen text in video ocr_text_extract
Erase hard subtitles from video subtitle_removal
Add/embed subtitles to video (burn-in subtitles) add_subtitle
Video scene segmentation/intelligent slicing intelligent_slicing
Portrait matting/portrait segmentation portrait_matting
Green screen keying/green screen chroma keying green_screen
AI comic drama conversion (comic style/3D cartoon style) comic_style
Short drama highlight editing/exciting clip extraction highlight
AI Video Translation (Subtitle/Voice/Face Translation) video_translation
Query Translation Project Status / Restart Translation Polling poll_translation
Query Translation Project List list_translation
AI Commentary Video Generation (Short Drama Commentary/Derivative Creation) drama_recap
AI Script Restoration (Video to Structured Script) drama_script
Query Media Asset Information (Vid Details + Playback Address) get_media_info

3) Construct Parameters and Execute

Video Editing Category

Script Purpose Detailed Parameters
stitching.py '<json>' Video/Audio Concatenation references/01-stitching.md
clipping.py '<json>' Video/Audio Trimming references/02-clipping.md
flip.py '<json>' Video Flipping references/03-flip.md
speedup.py video '<json>' Video Speed Change references/04-speedup.md
speedup.py audio '<json>' Audio Speed Change references/04-speedup.md
image_to_video.py '<json>' Image to Video references/05-image-to-video.md
compile.py '<json>' Audio-Video Synthesis references/06-compile.md
extract_audio.py '<json>' Extract Audio Track references/07-extract-audio.md
mix_audios.py '<json>' Audio Mixing references/08-mix-audios.md

Media Processing Category

Script Purpose Detailed Parameters
voice_separation.py '<json>' Vocal Separation references/10-voice-separation.md
noise_reduction.py '<json>' Audio Noise Reduction references/11-noise-reduction.md
quality_enhance.py '<json>' Comprehensive Image Quality Restoration references/12-quality-enhance.md
super_resolution.py '<json>' AI Super-Resolution references/13-super-resolution.md
interlacing.py '<json>' Intelligent Frame Interpolation references/14-interlacing.md

AI Content Analysis Category

Script Purpose Detailed Parameters
asr_speech_to_text.py '<json>' Speech Recognition ASR references/15-asr-speech-to-text.md
ocr_text_extract.py '<json>' OCR Text Extraction references/16-ocr-text-extract.md
subtitle_removal.py '<json>' Hard Subtitle Removal references/17-subtitle-removal.md
add_subtitle.py '<json>' Add Embedded Subtitles references/18-add-subtitle.md
intelligent_slicing.py '<json>' Intelligent Scene Segmentation references/19-intelligent-slicing.md
portrait_matting.py '<json>' Portrait Matting references/20-portrait-matting.md
green_screen.py '<json>' Green Screen Keying references/21-green-screen.md
highlight.py '<json>' Short Drama Highlight Editing references/23-highlight.md
get_media_info.py '<json>' Media Asset Information Query references/27-get-media-info.md

AI Content Generation Category

Script Purpose Detailed Parameters
comic_style.py '<json>' AI Comic Drama Style Transfer references/22-comic-style.md
video_translation.py '<json>' AI Video Translation references/24-video-translation.md
drama_recap.py '<json>' AI Commentary Video Generation references/25-drama-recap.md
drama_script.py '<json>' AI Script Restoration references/26-drama-script.md

Restart Polling

Script Purpose
poll_vcreative.py <task_id> Restart Editing Category Task Polling
poll_media.py <task_type> <RunId> Restart Media Processing Category Task Polling
poll_translation.py <ProjectId> Restart Translation Task Polling

In the timeout response, the resume_hint.command field contains a restart command that can be copied and executed directly.

Example

# 本地文件先上传拿到 vid（后续脚本统一用 vid://... 作为输入）
python <SKILL_DIR>/scripts/upload_media.py "/path/to/local.mp4" my_space

# 拼接两个视频，加转场
python <SKILL_DIR>/scripts/stitching.py \
  '{"type":"video","videos":["vid://v0001","vid://v0002"],"transitions":["1182359"]}'

# 使用 @file.json 传参（推荐，避免转义问题）
python <SKILL_DIR>/scripts/stitching.py @params.json

# 人声分离（注意 type 首字母大写）
python <SKILL_DIR>/scripts/voice_separation.py '{"type":"Vid","video":"v0310abc"}'

# 超分到 1080P
python <SKILL_DIR>/scripts/super_resolution.py '{"type":"Vid","video":"v0310xyz","Res":"1080p"}'

# ASR 语音识别
python <SKILL_DIR>/scripts/asr_speech_to_text.py '{"type":"Vid","video":"v0310abc"}'

# 短剧高光剪辑
python <SKILL_DIR>/scripts/highlight.py '{"Vids":["v023xxx","v024xxx"]}'

# AI 视频翻译（中文→英文）
python <SKILL_DIR>/scripts/video_translation.py '{"Vid":"v0d225gxxx","SourceLanguage":"zh","TargetLanguage":"en"}'

# AI 漫剧转绘（漫画风 720p）
python <SKILL_DIR>/scripts/comic_style.py '{"Vid":"v0d012xxxx","Style":"漫画风","Resolution":"720p"}'

# AI 解说视频（自动生成解说词）
python <SKILL_DIR>/scripts/drama_recap.py '{"Vids":["v023xxx"],"AutoGenerateRecapText":true}'

# AI 剧本还原
python <SKILL_DIR>/scripts/drama_script.py '{"Vids":["v023xxx","v024xxx"]}'

# 查询媒资信息
python <SKILL_DIR>/scripts/get_media_info.py '{"vids":"v001,v002"}'

# 超时后重启编辑类轮询
python <SKILL_DIR>/scripts/poll_vcreative.py <异步智剪任务ID> my_space

# 超时后重启媒体类轮询
python <SKILL_DIR>/scripts/poll_media.py videSuperResolution run_yyy my_space

# 超时后重启翻译轮询
python <SKILL_DIR>/scripts/poll_translation.py <ProjectId> my_space

Error Output

All errors use a unified format:{"error": "说明"}

Timeout Output (Including Restart Command):

{
  "error": "轮询超时（360 次 × 5s），任务仍在处理中",
  "resume_hint": {
    "description": "任务尚未完成，可用以下命令重启轮询",
    "command": "python <SKILL_DIR>/scripts/poll_media.py videSuperResolution run_yyy my_space"
  }
}

Constraints

  • Before calling the scriptmustview the detailed script parameter description

Billing Description

Only when the user proactively asks about fees or billing rules, refer to the billing description in references/00-billing-instructions.md and briefly explain to the user the billing composition of the VOD resources that volcengine-ai-mediakit depends on, avoiding proactively expanding on billing details in ordinary editing/processing conversations.