Generate audio documentaries from web articles with ElevenLabs and BrowserAct

Go to Workflow
1 views
Built by Madame AI Team | Kai Madame AI Team | Kai
Created on June 06, 2026

Description

Generate audio documentaries from web articles to Telegram with ElevenLabs & BrowserAct

This workflow transforms any web article or blog post into a high-production-value audio documentary. It automates the entire production chain—from scraping content and writing an engaging narrative script to generating realistic voiceovers—delivering a listenable MP3 file directly to your Telegram chat.

Target Audience
Commuters, podcast enthusiasts, content creators, and researchers who prefer listening to content over reading.

How it works
Analyze Intent: The workflow receives a message via Telegram. An AI Agent (using Google Gemini) classifies the input to determine if it is a casual chat or a request to process a URL.
Scrape Content: If a valid link is detected, BrowserAct executes a background task to visit the webpage and extract the raw text.
Write Script: A Scriptwriter Agent (using Claude via OpenRouter) converts the dry article text into a dramatic, narrative-driven script optimized for audio, including cues for pacing and tone.
Generate Audio: ElevenLabs synthesizes the script into high-fidelity speech using a specific voice model (e.g., "Liam").
Deliver Output: The workflow sends the generated MP3 file and a formatted HTML summary caption back to the user on Telegram.

How to set up
Configure Credentials: Connect your Telegram, ElevenLabs, OpenRouter, Google Gemini, and BrowserAct accounts in n8n.
Prepare BrowserAct: Ensure the AI Summarization & Eleven Labs Podcast Generation template is saved in your BrowserAct account.
Select Voice: Open the Convert text to speech node and select your preferred ElevenLabs voice model.
Configure Model: Open the OpenRouter node to confirm the model selection (e.g., Claude Haiku) or switch to a different LLM for scriptwriting.
Activate: Turn on the workflow and send a link to your Telegram bot to test it.

Requirements
BrowserAct* account with the *AI Summarization & Eleven Labs Podcast Generation** template.
ElevenLabs** account.
OpenRouter** account (or access to an LLM like Claude).
Google Gemini** account.
Telegram** account (Bot Token).

How to customize the workflow
Change the Persona: Modify the system prompt in the Scriptwriter node to change the narrative style (e.g., from "Documentary Host" to "Comedian" or "News Anchor").
Switch Output Channel: Replace the Telegram output node with a Google Drive or Dropbox node to archive the generated audio files for a podcast feed.
Multi-Voice Support: Add logic to split the script into multiple parts and use different ElevenLabs voices to simulate a conversation between two hosts.

Need Help?
How to Find Your BrowserAct API Key & Workflow ID
How to Connect n8n to BrowserAct
How to Use & Customize BrowserAct Templates

Workflow Guidance and Showcase Video

How to Build an AI Podcast Generator: n8n, BrowserAct & Eleven Labs

Nodes Used (5)

AI Agent
@n8n/n8n-nodes-langchain.agent
Google Gemini Chat Model
@n8n/n8n-nodes-langchain.lmChatGoogleGemini
OpenRouter Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenRouter
Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured
Telegram
n8n-nodes-base.telegram