Telegram trigger with audio and text branches for ai agent at the same time doesn't work

[Usuljedi]

“Hello, when I set up a switch that splits audio into one branch and text into another, and then they merge in the AI Agent, it doesn’t work. However, if I use only one of them, it works and the Telegram message-sending node gets activated. I’ve seen all the tutorials and guides and nothing works. Both branches have a Set that converts both messages into json.text, but it seems that the agent gets stuck, even though in the YouTube tutorials it works for everyone.”

[n8n]

It looks like your topic is missing some important information. Could you provide the following if applicable.

• n8n version:
• Database (default: SQLite):
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• Running n8n via (Docker, npm, n8n cloud, desktop app):
• Operating system:

[Usuljedi]

@n8n

• n8n version:
• Database (default: SQLite):
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• Running n8n via (Docker, npm, n8n cloud, desktop app):
• Operating system:

n8n version: 1.77.3
Database: Postgres
n8n EXECUTIONS_PROCESS: main
Running n8n via: n8n cloud
Operating system: windows and n8n Cloud

[Ghassan_Ibrahim]

Hey! I was just facing the same issue and found a solution for it!! So excited to share it with you!!

Use this code in the prompt user of the AI agent: {{ $json.message.text || $json.text }} which means json.message.text OR json.text

Let me know if it works for you!