github
Voice Input for Prompts
Using voice prompts has become popular recently. The reason is that some believe speaking directly, rather than typing, preserves more of the user’s tone, which large language models can pick up on. This helps reduce misunderstandings that often occur with plain text input, resulting in more accurate expression of intent.
Is it possible to add this feature to Copilot? Please note, I’m referring to direct voice input that is interpreted by the model—including tone and inflection—not simply converting speech to text and then processing it, as that would be no different from entering text directly.
Hi @owenzhao, this is on our radar, and we plan to support it in the upcoming release.
