Video course · 9 chapters · 21 min · certificate
Build a Multimodal AI Agent: See, Speak, and Chat with n8n & Telegram
Build a no-code multimodal Telegram agent in n8n that chats, listens, speaks and sees: Telegram trigger, OpenAI brain, conversation memory, voice transcription, text-to-speech replies, routing by message type and image analysis with questions.
What you'll learn
- Set up a Telegram trigger in n8n
- Add an OpenAI agent with memory
- Transcribe voice messages
- Reply with generated speech
- Route text, voice and images
- Analyse images and answer questions
Chapters
9 chapters · 21:12-
1:40
01Intro Free preview
Demo and overview
An agent that sees and speaks.
-
1:12
02Setup Members
Telegram trigger and OpenAI brain
Trigger and model.
-
1:31
03Memory Members
Conversation memory
Remembering context.
-
1:37
04Voice Members
Voice message processing
Receiving voice notes.
-
3:30
05Voice Members
Text-to-speech replies
Speaking back.
-
1:53
06Routing Members
Routing by message type
IF conditions.
-
6:11
07Vision Members
Image analysis
Seeing images.
-
1:49
08Vision Members
Asking questions about images
Using captions.
-
1:49
09Wrap-up Members
Final results
Testing all modes.
Jobs this course suits
Our AI checked this course against 500 jobs; these get the most out of it. Each job links to its learning path.