Media
2 items
Description
Dictate short text into supported web fields with a visible microphone button and secure backend transcription. Use your voice to write short messages, notes, searches, and form entries directly in Chrome. Dictozy adds voice dictation to supported text fields. Focus a field, click the visible microphone button or use your assigned browser shortcut, speak, and stop recording. Dictozy sends the clip over HTTPS to its backend for speech-to-text transcription, then inserts the returned text where you were writing. What you can do - Use voice typing in normal text inputs, textareas, and supported rich-text fields. - Start and stop recording from a clear microphone control beside the field. - Start, stop, or cancel dictation with a configurable browser shortcut. - Cancel a pending transcription and record again after a failure. - Preserve the intended caret or selection in supported modern forms. - Choose Automatic or one of 25 explicit language-formatting options. - Adjust the recording limit from the extension popup. - Turn the extension off globally or disable it only for the current website. Privacy and control - Recording begins only after you click the microphone button or use the assigned shortcut. - You can stop recording immediately. - Audio is sent to the Dictozy backend only for transcription. - The backend calls xAI Speech-to-Text; the extension never calls xAI directly. - The xAI API key remains on the backend and is never included in the extension. - Password, payment, hidden, disabled, readonly, file, checkbox, and radio fields are ignored. - Sites are enabled by default. Only exact origins you explicitly disable are stored locally, and those preferences are never sent to the backend. - The extension does not provide transcript history, advertising, sign-in accounts, payments, or background recording. Supported fields include text, search, email, URL, and telephone inputs; textareas; contenteditable fields; and ARIA textboxes that are actually editable. Bare non-editable ARIA textbox roles are ignored. The product is designed for short dictation clips rather than long-form recording. Language formatting is a provider formatting hint, not a promise of improved recognition. English is selected by default, while Automatic leaves provider language and formatting selection unset. What's new in 0.1.9 - Keeps insertion anchored to the caret or selection captured when recording starts. - Preserves surrounding rich-text formatting while inserting transcript text as plain text. - Avoids duplicated spaces around inserted transcripts. - Continues to reject late insertion if focus moves or the original field becomes unavailable.