Problem Summary:
Currently, there are two issues limiting the flexibility of speech-to-text in the Agent typing bar:
-
Speech-to-Text Disabled After Image Paste
-
When users paste an image into the Agent typing bar, speech-to-text becomes unavailable
-
Current workaround: Users must inject text via speech-to-text before pasting images
-
Desired behavior: Speech-to-text should remain active after pasting images
-
-
Cannot Continue Speech Input
-
After injecting text via speech-to-text and pressing the microphone button again, users cannot continue adding text
-
Desired behavior: Users should be able to press the microphone multiple times consecutively to keep adding speech-injected text to the Agent typing area
-
Impact:
Many users rely on speech-to-text as their primary method for prompt injection while working with Agents. These limitations reduce workflow efficiency and flexibility.
Feature Request:
Enable speech-to-text to work seamlessly alongside image pasting and allow multiple consecutive speech-to-text inputs without interruption.

