Turn a Telegram chat into a smart content and reply assistant. Users send text or voice notes, and the system answers fast, saves helpful notes, and can draft posts or blogs on request. It fits marketing and community teams that want quick replies and reusable content ideas.

Here is how it works. A Telegram trigger captures each message. A switch routes text directly, or pulls the voice file and sends it to OpenAI to transcribe. The text from either path is combined with past notes loaded from Airtable, which are grouped and merged into one context. An AI agent uses this context with a chat model and short term memory, can look up facts with Wikipedia, do quick math, write new memory back to Airtable, and even call a separate content creator workflow. The reply is sent back to Telegram.

Setup needs a Telegram bot token, an OpenAI key, and an Airtable Personal Access Token with a base and table for memory. Expect faster reply times, less copy and paste, and a growing memory of user details. Good for education channels, product communities, and content teams turning chats and voice notes into ready drafts.

• Telegram trigger listens to new chat messages and voice notes
• Switch node routes text and voice into the right path
• Telegram file node grabs the voice file for processing
• OpenAI transcribes voice to text for fast responses
• Airtable search loads up to 50 memory rows and Aggregate groups them
• Merge combines current text and past memory into one context
• AI agent uses an OpenAI chat model and Simple Memory per chat session
• Tools include Wikipedia and Calculator for facts and quick math
• Airtable tool writes new memory and a content creator tool drafts posts or blogs

• Reduce Telegram reply time from 10 minutes to 1 minute
• Automate up to 70 percent of routine chat answers
• Improve note capture accuracy by 90 percent with Airtable records
• Handle 5 times more chat volume without adding headcount
• Connect Telegram, OpenAI and Airtable in one flow

1. Import the template into n8n: Create a new workflow in n8n, click the three dots menu, select 'Import from File', then choose the downloaded JSON file
2. You'll need accounts with Telegram, OpenAI, Airtable and Google Gemini. See the Tools Required section above for links to create accounts with these services.
3. Create a Telegram bot with BotFather and copy the bot token. In n8n, open the Telegram Trigger node, choose Credential to connect with, click Create new credential, and paste the token. Use the same credential on both Telegram nodes.
4. In the Telegram Trigger node, keep Updates set to message. Make sure the bot is added to your chat or group so messages reach the workflow.
5. Generate an OpenAI API key in your OpenAI account. In n8n, open the OpenAI nodes, create a new OpenAI credential, and paste the API key. Select this credential for both the audio transcribe and chat model nodes.
6. Create an Airtable Personal Access Token with read and write scopes. In n8n, open the Airtable nodes, create a new Airtable credential, and paste the token. Confirm the Base and Table match your memory base and Memory table.
7. Check the Airtable table has a text field named Memory so records can be searched, aggregated, and created by the memory tool.
8. (Optional) Create a Google Gemini API key in Google AI Studio. In n8n, open the Google Gemini Chat Model node and add the Gemini credential. This model can be used if you switch the AI agent model later.
9. Open the AI Agent node and review the system message and tools. Adjust the tone or instructions to match your brand, and keep the tools enabled for memory, Wikipedia, and content creation.
10. Verify the Merge node receives text from either the Edit Fields path or the OpenAI transcription path, and also receives the aggregated Airtable memory. Save the workflow.
11. Run once in test mode and send a Telegram text and a voice note to confirm replies and Airtable memory writes. When stable, set the workflow to Active.