SM
Sheby M
Loading portfolio experience
Business Operations, Marketing · AI Agent, Automation, Dashboard
PodArchive AI – YouTube Podcast Transcription Library
An automated transcription workflow that discovers authorized podcast episodes on YouTube, generates searchable transcripts, and organizes every episode into structured Google Docs.
Stack and focus areas
AutomationVoice AIAIPythonOpenAI WhisperYouTube Data APIGoogle Drive APIGoogle Docs APISpeech-to-TextAudio ProcessingSpeaker DiarizationTimestamp GenerationTranscript FormattingEpisode DiscoveryScheduled WorkflowsMetadata ExtractionQuality ReviewEntertainmentProfessional Services
Project description
Creating a complete research library for podcasts such as Home Service Millionaire, Home Service Expert, My First Million, EntreLeadership, Small Marketing Big Business, Franchise Secrets, and Service Business Mastery required discovering every episode, processing hours of audio, formatting transcripts, and organizing them into separate master documents.
I developed an automated transcription pipeline that uses the YouTube Data API to discover episodes from configured channels and playlists. For content authorized for processing, the workflow extracts episode metadata and audio before using OpenAI Whisper to generate timestamped transcripts. The text is cleaned, divided into readable sections, and optionally separated by speaker. Each transcript is added to the corresponding podcast’s Google Drive folder and master Google Doc with its title, publication date, video link, and processing details.
