Written for candidates preparing for CMS or WSET tasting examinations.
The tasting exam is spoken as much as it is written — you narrate the glass. Dictation + Transcription lets a practice session capture that narration and hand it back as a transcript; Voice Control goes further and lets a CMS Advanced/Master session fill the grid from your speech, hands-free. Both come with the CMS Advanced/Master plan — see Plans & What They Include.
Dictation + Transcription
Enable audio recording when setting up a blind tasting, choose your microphone, and speak the tasting as you work — the running commentary you'd give an examiner. Recording begins with the session's Start Timer press, runs without a pause control, and stops when you save — or at the thirty-minute mark, whichever comes first. Your audio is processed by a speech-recognition service to produce the transcript, and when you save, both attach to the note's Dictation & Transcript panel: the recording with a player, the transcript synced to it, so a click in the text seeks the audio.
The study value is in the replay: what you said about the wine against what you concluded, and both against what the wine turned out to be. Verbal tics, hedges, and the aromas you named aloud but never entered — the transcript keeps them all, and they're often the most instructive part of the review.
Recording requires the browser's microphone permission, requested on first enable; deny it and both features stand down until it's granted. The audio itself uploads just after the save completes, so give the page a moment before closing the browser; should an upload fail, the transcript still saves with the note.
Voice Control
In CMS Advanced/Master sessions with recording enabled, Voice Control listens for the grid itself: speak your calls and the format fills hands-free — eyes on the glass, hands on the stem, the way the real examination runs. It's available exactly there — Advanced/Master format, recording on — and nowhere else, by design.
How to Phrase a Call
The listener keys on field names and values, and a few patterns carry most of the grid:
Descriptors and aromas can surface anywhere in a sentence — "on the nose, the fruit is candied and jammy" checks Candied and Jammy.
Single-choice fields want the field's name with the value: "Oak Origin is French," "the wine is balanced." Add "not" to select No: "the wine is not balanced."
Ambiguous values need their field named: "the primary color is ruby," "there is blue fruit on the nose."
Sliders answer to the field plus a point on its scale: "color intensity is deep," "the complexity is medium plus."
One value at a time in the conclusion fields — "possible varieties include Cabernet Sauvignon" lands; chaining several in one breath selects only the first.
A few entries stay manual by design: the main-characteristics free text, the region and appellation search, and the vintage year are typed or tapped, not spoken.
Speaking for the Microphone
Minimize background noise, and keep the path from mouth to microphone clear.
Enunciate, at a moderately quick pace, oriented toward the microphone.
Run a multi-word command as one breath, and leave a beat between back-to-back commands.
An external microphone can help in an echoing room; any decent input works.
Devices & Browsers
Both features are web-app capabilities: they run in desktop and tablet browsers. The native apps hand blind tastings to the web app, so any device lands in the right place on its own — candidates who practice with dictation should simply favor the roomier screens.
Privacy Posture
Recordings and transcripts are yours: they live with your notes, appear nowhere publicly, and are removed with your account if you ever delete it. For the governing terms, see the Privacy Policy.
Curate is an independent software provider, not affiliated with the Court of Master Sommeliers or the Wine & Spirit Education Trust, nor endorsed by either.