Video/Audio Transcription Skill

Overview

This skill provides automated transcription for video and audio content using Gladia API. It converts spoken content from YouTube videos, MP3 files, or any accessible audio/video URL into text, which can then be summarized by an LLM.

Use Cases

• YouTube Video Summary - Transcribe YouTube videos for LLM summarization (especially useful for Cantonese/Chinese content without captions)
• MP3/WAV to Text - Convert audio files to transcript
• Video Content Extraction - Extract speech from any publicly accessible video URL
• Podcast Transcription - Convert podcast episodes to text

Service: Gladia API

What is Gladia?

Gladia is an audio transcription API that supports multiple languages including Cantonese. It provides both async (pre-recorded) and real-time transcription.