Long-Form Transcript Analysis Assistant

Long-Form Transcript Analysis Assistant Overview This workflow automates the process of transcribing and analyzing long-form audio files. It listens for new audio files added to a specific Google Drive folder, sends an approval request to a user, and upon approval, uses AI to transcribe the audio. It then generates multiple report formats (structured JSON, detailed Markdown, and raw text), saves them back to Google Drive, and notifies the user via both Gmail and Telegram with links to the new documents. Technologies Used - n8n: The automation platform coordinating the entire workflow. - Google Drive: Used for storing the initial audio files and the final generated reports. - Gmail: Used for the "human-in-the-loop" approval step and for sending the final report notifications. - OpenAI: The core AI engine used for audio transcription (Whisper) and for summarizing and formatting the transcript into various structured outputs (GPT-4o-mini). - Telegram: Used for sending a secondary, concise notification with links to the reports. What It Automates This workflow automates the entire lifecycle of audio content processing, from ingestion to analysis and distribution. It eliminates the manual, time-consuming tasks of: 1. Downloading audio files. 2. Uploading them to a transcription service. 3. Manually summarizing key points, action items, and topics from the transcript. 4. Formatting summaries into different document types (JSON, Markdown). 5. Organizing and saving all related files. 6. Drafting and sending notification emails or messages. The inclusion of an approval step ensures that resources (and API costs) are only used when a user explicitly confirms the action, adding a layer of control and cost management. AI Integration Analysis Artificial Intelligence is central to this workflow, transforming unstructured audio into structured, actionable information. It is integrated at several key stages: 1. Audio Transcription: The Transcribe with OpenAI node uses OpenAI's Whisper model to convert the raw audio file into a text transcript. This is the foundational step that turns spoken words into machine-readable data. 1.