Ai Agent can not get text from last node

[J_oseph]

I have a strange question. I’m working on a small LINE + AI agent bot. So far, feeding text and images to the AI works fine. However, when voice input comes in, I use Google Speech-to-Text to convert it into text, but the resulting text somehow doesn’t get picked up by the AI.
Strangely though, if I click in and view the generated text, it displays correctly.

Here’s the ternary expression I’m using:

{{ 
  $json.body.events[0].message.type == "text" 
    ? $json.body.events[0].message.text 
    : ($json.body.events[0].message.type == "image" 
        ? "Please check my image" 
        : ($json.body.events[0].message.type == "audio" 
            ? $('AudioToText').item.json.results[0].alternatives[0].transcript 
            : "error"))
}}

Information on your n8n setup

• n8n version: 1.88.0
• Database (default: SQLite): PostgreSQL
• n8n EXECUTIONS_PROCESS setting (default: own, main): Default
• **Running n8n via (Docker, npm, n8n cloud, desktop app):**Docker on k8s
• Operating system: k8s with ubuntu 22

[ThinkBot]

You could just try reference each previous variable without making a conditional waterfall. That may help. I am unsure why you are encountering the issue tho.

[J_oseph]

But other two flow is workable