Forum Discussion
Copilot Image Generation Tool Fails When Verbal Response Comes First
Hi everyone,
I’ve encountered a recurring issue with Copilot in Windows. When Copilot responds verbally before triggering the image generation tool (e.g., “Here’s your image” or “Let me show you”), the image tool often fails to activate. There’s no error message, and the usual visual cue (a small rectangle showing the progress of the image generation beneath the message) doesn’t appear. The image simply isn’t generated.
I’ve been using Copilot for about 2 weeks, and this issue has happened consistently. If Copilot skips the verbal preamble and jumps straight into generating the image, it works fine.
I’ve already submitted feedback through support, but I’m posting here in case others have experienced this or have insights. Happy to provide examples or help test further.
Thanks!
3 Replies
Follow-Up: Bug Affects All Output Types - But Not in Think Deeper Mode
Hey again, just following up on the issue I posted earlier about Copilot not generating content after saying things like “Let me show you” or “Here’s your image.” I’ve since realized this isn’t limited to images,it affects any kind of output, including quizzes, summaries, and charts.
For example, I asked Copilot to make me a quiz. It replied:
“Here's a general knowledge quiz to get you started!”
But then… nothing. No quiz ever appeared.Here’s a screenshot of that moment:
Copilot says it's preparing a quiz, but no quiz appears afterward.
New Clue: I’ve discovered that this issue doesn’t happen when I switch the chat mode to Think Deeper. In that mode, Copilot takes a few extra seconds to respond but the output actually shows up. So it seems like the bug is tied to how quickly Copilot tries to respond in default mode, possibly cutting off the second part of its reply.
Platform: Windows 11 app
Mode: Text chat
Frequency: Happens in default mode, not in Think Deeper
Impact: Breaks user flow and makes Copilot appear unresponsiveStill haven’t gotten any replies, so just bumping this again. Would really appreciate any insight or confirmation from others—or even just knowing it’s on someone’s radar.
Thanks!
Thanks for the suggestion! Just to clarifythis isn’t about Copilot speaking out loud via the voice feature. The issue happens even when voice is disabled.
What I’m referring to is when Copilot writes a verbal intro in the chat, like:
- “Let me create that for you…”
- “Here’s what I came up with:”
- “Generating your image now…”
- “This should match what you described:”
- “I’ve made something based on your prompt:”
In those cases, the image tool often doesn’t activate, and no image appears. But when Copilot skips the intro and jumps straight into generating the image, it works fine.
So it’s a text-based behavior, not an audio setting. Hope that clears it up!
Disable Voice Responses Temporarily
Go to Settings > Accessibility > Speech and toggle off "Speak answers aloud" (or similar, depending on your Windows version).
Test if image generation works more reliably without voice.
