Speech to text happens in your browser
Transcription uses the Web Speech API built into Chrome and Edge, so your microphone audio never leaves your device at this step — only the text is sent to generate the new voice. Speak in the language of the voice you've selected; the recognizer is set to match it. Firefox and Safari don't expose this API reliably, so on those browsers paste the text instead. Check the transcript before generating: a wrong word in the text becomes a wrong word in the audio.