Audio Summary Generator: A Workflow Script Tutorial

User provides a topic. The script researches it on Google, writes a concise summary, then converts that summary to an audio file.

Why build this

This tutorial introduces chaining three actions inside a single script — Search Google, Generate Text, and Text to Speech. You’ll see how each action’s output feeds into the next: search results become source material for a written summary, and that summary becomes the input for the Text to Speech action. The final output is an audio file URL that the workflow can pass to the next block.

Blocks you’ll use

• User Input — Collects the topic the user wants summarized and converted to audio.
• Run Script — Houses the Workflow Script. Inside the script, a Search Google action researches the topic, a Generate Text action writes a short summary, and a Text to Speech action converts it to audio.
• Display Content — Shows the generated audio file to the user.

Canvas flow

Start → User Input → Run Script → Display Content → End

The Run Script block is where your Workflow Script lives. It chains three actions — Search Google, Generate Text, and Text to Speech — so the search results feed into the summary, and the summary feeds into audio conversion. The final audio URL is returned under the key audio.

How to build it

1. Open the Automations Canvas — Open MindStudio. The Automations Canvas is displayed by default. You’ll see a Start block and an End block already in place.

2. Add a User Input block — Click + to add a block and select User Input. In the block settings on the right, click + to add a user input, then click Create New. Set the Type to Short Text and enter topic in the Variable field — this is the name you’ll use to reference this input in the Run Script block. Set the Label to a short name — for example, “Topic”. Then set the Help Text to a description the user will see — for example, “What topic should I summarize?”

   Once you’ve set up the user input, click the Main.flow tab at the top of the editor to return to the workflow.

3. Add a Run Script block — Click + in the workflow, then select View All Blocks. From the Blocks screen, search for Run Script and select it. In the block settings, click Edit Script to open the script editor. Paste the prompt below to generate the script automatically.

   Generate a script that takes a topic as input. Use a Search Google action to research the topic, then use a Generate Text action to write a concise 2–3 sentence summary. Then use a Text to Speech action to convert the summary to audio. Return the audio URL under a key named audio.

   • Return to the workflow — Once the script is generated, click the Main.flow tab to return to the workflow. The Input Variables and Output Variables settings will appear in the block settings on the right.
   • Configure Input Variables — For the topic parameter, set the Variable to {{topic}}.
   • Configure Output Variables — Confirm that audio is mapped to the variable {{audio}}.

4. Add a Display Content block — Click + and select Display Content. Connect it after the Run Script block. In the block settings, set the Message field to <audio src="{{audio}}" controls /></audio> — using an HTML audio tag to render the audio URL returned by the script as a playable audio player. This displays the generated audio to the user.

5. Name your AI agent — Click Untitled Agent at the top of the interface, then update the Agent Name field — for example, “Audio Summary Generator.”

6. Preview your AI agent — Click the Preview button, then select Open Draft Agent to run the AI agent and test your workflow.

Try it in MindStudio

Open MindStudio and follow along with this tutorial. No coding required.