Who is this workflow for? This workflow automates the transcription and analysis of sales and support calls using AssemblyAI and OpenAI, storing results in Supabase. It streamlines the process of extracting actionable insights from recorded conversations, enhancing team performance and client interactions..

What does this workflow do?

1. Transcribe Call with AssemblyAI

  • Set Up Request:
  • Authentication: Include the AssemblyAI API key in the Authorization header.
  • Endpoint: POST to https://api.assemblyai.com/v2/transcript/.
  • Parameters:
    • audio_url: Direct link to the audio file stored in Supabase or Dropbox.
    • webhook_url: URL for an n8n webhook to receive transcription results.
    • Additional Settings:
    • speaker_labels: Enable speaker diarization (true or false).
    • speakers_expected: Specify the number of speakers.
    • language_code: Set to the desired language, default is en_us.

2. Process Transcription with OpenAI

  • Webhook Configuration: Set up a POST webhook in n8n to receive transcription data from AssemblyAI.
  • Retrieve Transcription:
  • Authentication: Use the AssemblyAI API key in the Authorization header.
  • Endpoint: GET https://api.assemblyai.com/v2/transcript/<transcript_id>.
  • Send to OpenAI:
  • Endpoint: POST to https://api.openai.com/v1/chat/completions.
  • Authentication: Include the OpenAI API key in the Authorization header.
  • Body Parameters:
    • model: Choose gpt-4o-2024-08-06 for full JSON schema support or gpt-4o-mini for cost efficiency.
    • messages:
    • system: Contains the analysis prompt guiding the AI.
    • user: Provides the combined text of the speakers’ utterances.
    • Response Format:
    • type: Set to json_schema.
    • json_schema: Define the schema for structured AI responses.
  • Store Results in Supabase:
  • Operation: Create a new record in the Supabase database.
  • Table Name: demo_calls.
  • Fields:
    • Input: Includes transcription text, audio URL, and transcription ID.
    • Output: Contains the parsed JSON response from OpenAI’s analysis.

🤖 Why Use This Automation Workflow?

  • Automated Transcription: Converts audio calls to text with speaker identification, eliminating manual transcription efforts.
  • AI-Driven Analysis: Extracts key metrics such as client intent and upsell opportunities, providing deeper understanding of conversations.
  • Centralized Data Storage: Stores transcriptions and analysis results in Supabase for easy access and integration with other tools.

👨‍💻 Who is This Workflow For?

This workflow is designed for:

  • Sales Teams: Enhance sales strategies by analyzing client interactions.
  • Customer Support Managers: Improve support quality through detailed call analysis.
  • Online Education Services: Gain insights from follow-up calls to better address student needs and opportunities.

🎯 Use Cases

  1. Sales Performance Improvement: Analyze sales calls to identify successful tactics and areas for improvement.
  2. Customer Support Optimization: Review support interactions to enhance service quality and customer satisfaction.
  3. Educational Feedback Analysis: Assess follow-up calls with students to tailor educational offerings and support services.

TL;DR

This workflow seamlessly integrates AssemblyAI for transcribing calls and OpenAI for analyzing the transcriptions, with Supabase providing a robust storage solution. By automating these processes, teams can efficiently derive meaningful insights from client interactions, enhancing sales performance, customer support, and educational services.

Help us find the best n8n templates

About

A curated directory of the best n8n templates for workflow automations.