skills.rest
quanpersie2001
quanpersie2001COMMUNITY
Explore all skills by this author →
GitHub

ai-multimodal

Process multimodal media to extract text, transcriptions, and structured data.

Updated Mar 27, 2026
Search Tags:#ocr#image generation#multimodal#transcription#gemini#video generation#document extraction
One-click install
npx skills add https://github.com/quanpersie2001/pulse --skill ai-multimodal-quanpersie2001
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-multimodal
Source: https://github.com/quanpersie2001/pulse/tree/main/plugins/pulse/skills/ai-multimodal
Command: npx skills add https://github.com/quanpersie2001/pulse --skill ai-multimodal-quanpersie2001

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, pypdf, python-docx, docx2pdf, markdown, Pillow, python-dotenv, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Unlocks advanced multimodal processing so teams can reliably extract, transcribe, analyze, and generate content from audio, images, video, and documents without manual pre-processing or ad hoc tooling. It prevents ad-hoc workflows that produce unlabeled outputs, missing timestamps, or undocumented transformations by providing opinionated scripts and reference guidance for robust media handling.

Core Features & Use Cases

  • OCR & Document Extraction: Convert PDFs and scanned images into clean Markdown and structured outputs for archival or review.
  • Audio & Video Transcription: Produce timestamped transcripts, speaker labels, and summaries from long recordings with chunking and retries.
  • Vision & Analysis: Captioning, object detection, visual Q&A, and screenshot extraction for product QA or content verification.
  • Image & Video Generation: Produce images (Imagen/Nano Banana) and short Veo videos with model fallbacks and billing-aware error handling.
  • Operational Tooling: Bundled CLI scripts for preflight (media_optimizer), batch processing (gemini_batch_process), document conversion, and setup checks with API key rotation support and optional ffmpeg/Pillow optimizations.

Quick Start

Transcribe and summarize meeting.mp3 with speaker labels, timestamps, and a markdown-formatted summary using the bundled CLI.

Frequently Asked Questions about ai-multimodal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract text and structured data from PDFs and scanned images?▼

Extract text from PDFs and scanned images using OCR and document extraction scripts that convert content into clean Markdown or structured JSON outputs. This handles archival or review workflows reliably without manual pre-processing.

Can I transcribe audio recordings with timestamps and speaker labels?▼

Transcribe audio recordings with timestamped transcripts, speaker labels, and summaries using bundled processing scripts. These scripts handle long recordings by applying chunking and retries to ensure complete transcription.

Does this multimodal processing workflow support video generation?▼

Video generation is supported to produce short Veo videos with model fallbacks and billing-aware error handling. The workflow also supports image generation using Imagen and Nano Banana models.

How do I handle large file size uploads for Gemini multimodal processing?▼

Handle large file size uploads for Gemini multimodal processing using the File API for uploads and preflight optimization. The bundled CLI scripts apply ffmpeg and Pillow optimizations to prepare media assets before processing.

What is needed to run preflight media optimization for transcription workflows?▼

Preflight media optimization for transcription workflows requires setting up your environment with API keys, python-dotenv, and optional ffmpeg and Pillow dependencies. The bundled media_optimizer script checks this setup and applies necessary transformations.

Are there limitations when generating images with Nano Banana and Veo?▼

Generation limitations with Nano Banana and Veo include potential billing errors or model unavailability, which are handled via model fallbacks and billing-aware error handling to prevent workflow interruption during asset generation.

Related Skills

Explore More Software Engineering Skills→
dthuy62dthuy62

ck:ai-multimodal

Analyze media files and generate structured outputs via Gemini and MiniMax APIs.

Community
Advanced
alibabaalibaba
10.5k

pdf-extractor

Extract text, tables, and metadata from PDF documents into JSON.

Official
Intermediate
nguyenvanlinh1902nguyenvanlinh1902

ai-multimodal

Extract insights from images, audio, and video using Gemini multimodal APIs.

Community
Advanced
freekmurzefreekmurze
997

pdf

Extract text and tables from PDFs using pypdf and pdfplumber.

Community
Advanced