To manage different Python versions, we use pyenv. Follow the installation steps based on your operating system:

• On macOS:

  brew update
  brew install pyenv

• On Ubuntu:

  sudo apt update
  sudo apt install -y build-essential curl libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget curl llvm libncurses5-dev libncursesw5-dev xz-utils tk-dev libffi-dev liblzma-dev python-openssl git
  curl https://pyenv.run | bash

• On Windows:
  • Install pyenv-win via pip:

    pip install pyenv-win --target %USERPROFILE%\\.pyenv

  • Alternatively, use the manual installation method:

    git clone https://github.com/pyenv-win/pyenv-win.git "$HOME/.pyenv"

Add pyenv to your shell by updating your profile:

• For macOS and Linux:

  echo 'export PATH="$HOME/.pyenv/bin:$PATH"' >> ~/.bash_profile
  echo 'eval "$(pyenv init --path)"' >> ~/.bash_profile

• For Windows:
  • Add %USERPROFILE%\.pyenv\pyenv-win\bin and %USERPROFILE%\.pyenv\pyenv-win\shims to your system's environment variables under 'Path'.

To isolate our project dependencies, we'll set up a virtual environment using pyenv and pyenv-virtualenv. Follow these steps:

• Create a virtual environment:

  pyenv virtualenv venv

• Activate the virtual environment:

  pyenv activate venv

With your virtual environment activated, install the required Python packages using pip:

• Install packages:

  pip install -r requirements.txt

• Go to your RetellAI Dashboard
• Navigate to AI Agents, and click Add Agent
  • Select Connect Your Own Agent
  • You can rename the Agent, and choose a voice.
  • Note down the Agent ID.
    • This needed for the environment variable: RETELL_AGENT_ID later.
    • Agent ID will be specified when using Twilio calls.
  • Under Custom LLM URL, paste the LLM websocket url with /llm-websocket appended to the base URL.
    • Using ngrok: wss://abcd-12-34-56-789.ngrok.io/llm-websocket (See Obtain and Set Up ngrok)
    • Using fly: wss://pete-ai.fly.dev/llm-websocket

• Login to your Calendly account.
• On the left-hand navigation panel, select Integration & apps
• In the All integrations section, select API and webhooks
• Click Generate New Token
• Enter a name for your token
• Click on Copy Token. This will be the environment variable: CALENDLY_API_KEY

• Create a new project
  • Navigate to your Google Cloud Console
  • On the upper taskbar, next to the Google Cloud icon on the left-hand side, click the dropdown menu.
  • On the upper right-hand corner, click New Project, and create a new project.
  • Navigate to your project's dashboard by clicking the project name on the home page.
    • You can find this in the section under your name. With the text: You're working on project <your-project-name>
• Enable Google Calendar API
  • On the project dashboard, select APIs & Services on the left-hand panel.
  • On the search bar, search for Google Calendar. Select Google Calendar API then click Enable
  • Navigate back to your project's dashboard, then select APIs & Services, then select Credentials
  • On the upper taskbar, click CREATE CREDENTIALS then select OAuth client ID
    • Select External, then enter relevant information for your app such as App Name, and your Email
    • In the Scopes tab, select ADD OR REMOVE SCOPES, then select the ff scopes:
      • .../auth/userninfo.email
      • .../auth/userinfo.profile
      • openid
    • In the Test users tab, select ADD USERS, here you can add email addresses for test users.
    • Click Next and Continue then Back To Dashboard
  • On the left-hand navigation panel, navigate to Credentials, then click CREATE CREDENTIALS then select OAuth client ID.
    • Select Desktop app as application type.
    • A popup window will appear, confirming OAuth client has been created. In the popup window, click on DOWNLOAD JSON
    • Rename this file to credentials.json, and move it to the scripts folder in your project directory.
• Generating the Token
  • Run the file scripts/create_google_token.py. This will create a token.json file.
  • You will be redirected to google. Authenticate with your email, then click Continue
  • Copy the contents, this will be the value of the environment variable GOOGLE_TOKEN

• Copy the .env.example file to create a .env file:

  cp .env.example .env

• Manually edit the .env file and add your environment variables:
  • OPENAI_API_KEY=your-openai-api-key
  • RETELL_API_KEY=your-retell-api-key
  • RETELL_AGENT_ID=your-retell-agent-id
  • CALENDLY_API_KEY=your-calendly-api-key
  • GOOGLE_TOKEN=your-google-token

• Download and install ngrok from ngrok's website.
• Start ngrok with the following command to expose port 8080:

  ngrok http 8080

• Note the https URL that ngrok provides. You will need this URL to configure Slack to send events to your local machine.
• You can set the --subdomain argument to pass in a different subdomain if you're using a paid version of Ngrok.

• Enter the following command on the terminal to run the server

  uvicorn main:app --host 0.0.0.0 --port 8080

• Go to your RetellAI Dashboard and select your AI Agent.
  • For testing on the dashboard, we have to specify a voice.
  • Click the Voice Name under Agent ID.
  • Select Echo (Or you may choose any voice)
• Click Test Audio.
• You can now talk to your agent.