🎥How to Create Automations using Multiple AI Agents🗒️

Tess AI API

• API Documentation

• Allows integration of advanced AI functionalities in external workflows

• More robust than simple GPTs, allowing the combination of different AI models into a single agent

• Enables the creation of complex agents in a no-code manner in AI Studio

Example of an Agent for Audio Transcription

• Utilizes the Deepgram model, which outperforms OpenAI’s Whisper in Portuguese

• Accepts various audio formats as input (MP3, MP4, OGG, etc.)

• Generates output in JSON format for easy integration with other tools

• Can be customized to adapt the transcription to formal written language

Creating Agents in AI Studio

• Intuitive interface for creating complex agents without the need for programming

• Ability to add multiple steps such as:

  • Audio transcription

  • Web queries

  • Document processing (PDFs, spreadsheets)

  • Integration with external apps (Google Sheets, Drive)

• Capability to combine different AI models (GPT-4, Claude, Gemini) into a single agent

Integration with Automation Tools

• Demonstration of use with Zapier and Make (Integromat)

• Examples of workflows to process WhatsApp messages:

  • Audio transcription

  • Sentiment analysis of feedbacks

  • Generation of automatic responses

• Ability to create logs and store data in spreadsheets for later analysis

Practical Case - Feedback Classification

• Agent that analyzes feedback from restaurant customers

• Determines sentiment (positive, negative, neutral), category (food, delivery, service), and reason for the feedback

• Integration with Google Sheets for input and output of data

Automation for Conversational Documentation

• Workflow to process ZIP files exported from WhatsApp

• Extracts text and audio messages

• Automatically transcribes audio

• Generates a complete text file with the entire conversation, including transcriptions

• Facilitates the creation of knowledge bases for training AIs

Advantages of Tess AI

• Flexibility to combine multiple AI models into a single agent

• Easy integration with other tools and platforms via API

• Ability to create robust and customized agents for various use cases

• Capability to process and analyze large volumes of data

Complexity Levels and Costs

• Zapier: Simpler and no-code, but can be more expensive for large volumes

• Make (Integromat): More flexible than Zapier, good cost-benefit

• N8N: More technical, but offers more control and can be self-hosted

• Tess Orchestrator: More advanced and customizable solution, ideal for large volumes and complex cases

Discover all this at Tess AI!