The microphone button in the Ask box turns speech into text in the message box. It uses your browser's built-in speech recognition (the Web Speech API), only when you click it, and only in browsers that provide it.
What happens
- You click the microphone and your browser asks for permission.
- You speak. Your browser converts speech to text and puts it in the box.
- You edit the text if needed and send it like any other prompt.
Who hears your voice
In Chrome, Edge and Safari, the browser itself may send your voice to its vendor's speech service to produce the text. That is governed by your browser vendor's policy, not Keplar's. Keplar never receives, records or stores audio. It receives only the text in the message box.
When it is not available
In browsers without the feature, such as Firefox, the button is not shown. Microphone use is allowed only for Keplar's own origin and can be revoked in your browser's site settings.
Tips
Speak in full sentences, say punctuation words like "comma" if your browser supports them, and review the text before sending, because dictated numbers and names are easy to mishear.
Related
- How Keplar handles your data: Where a question goes, what is stored under your account, which companies process it, what is never sent to model providers, and what Keplar does not do.
- Write better questions: Eight habits that get more from a multi-model answer: state the decision, add constraints, ask for the counterargument, separate facts from judgment, and more, with prompts you can copy.