What a voice simulation is like for the learner
This article describes what a learner actually experiences in a voice simulation, so you know what you are sending them into. To turn voice on and configure it, see Set up a voice simulation.
The run is voice, from first frame to report
A voice simulation opens straight into the full-screen conversation. There is no text box, no message list and no way to type — the learner talks, the character answers out loud, and the run ends when the audio ends. The character speaks first: its opening line plays as soon as the call starts, so the learner does not have to break the silence. The transcript is still recorded, and the feedback report at the end is exactly the report a text simulation produces.
When the learner chooses text or voice
A creator can run a simulation as voice only, text only, or Let the learner choose. In the last case, before the run starts the learner sees How do you want to talk with followed by the character’s name, and picks Text chat or Voice call. The screen warns “Pick once. There is no switching midway.”
- On a computer, the two options are tiles with a preview of each.
- On a phone, they are two short rows, so both options and the start button fit on one screen.
Once the learner picks, the start button names the choice: Start in writing or Start by voice. On a page with more content further down, a More below button at the bottom edge scrolls to it.
What is on screen
Whichever experience the creator chose — Phone Call, Portrait or Video call — the same furniture is there:
- Who you are talking to — the character’s photo (or a silhouette, or their initial) and name.
- What is happening right now, in words: Connecting…, Listening, Speaking, Paused. Nobody has to work out from an animation whether the microphone is live.
- A live caption of the character’s last line, so a missed word does not derail the conversation. The learner can switch it with Show captions / Hide captions. A long line scrolls inside its own area rather than covering the controls, and keyboard users can focus the Captions area to scroll it. In a phone call, the dialer shows until the character first speaks, and then the captions take its place.
- Session time and Your messages — a countdown when the run has a time limit, a count-up when it does not.
- The controls: Pause / Resume, Advice, and End, always in the same place and always the last item in reading order.
Two tabs sit on the edge of the panel: View Intro reopens the briefing the learner read before starting, and Transcript shows the conversation so far. Pausing stops the clock as well as the microphone, so a learner can step away without spending their time limit.
Advice during the conversation
If the creator enabled the Realtime Guide, its advice is available in a voice run too — the Advice button opens it, and it changes to New advice when there is something fresh to read. The advice card carries the name of the simulation’s evaluator when it has one, so the coaching reads as coming from a person rather than from the system. See Enable and configure the Realtime Guide.
In a video call
If Cesura has set the simulation to a lifelike video avatar, the character appears as a lip-synced face in the call. See Video avatar mode.
The learner’s camera is off until they turn it on with Start video, and Stop video turns it off again. Nothing from the camera is transmitted or recorded — the picture stays on their own screen, so the exercise can involve being on camera without the platform keeping any of it. If their browser blocks the camera they are told how to allow it, and the call carries on with audio either way.
Long answers and short silences
The learner can speak at length. A long answer, such as a 20-second self-introduction, is not cut off mid-sentence. Pausing the call does not count as silence.
A call nobody speaks in pauses itself. After a stretch of silence the learner sees “Still there? Audio pauses in…” with a countdown, and then “Paused — no one was speaking. Resume when you’re ready.” Resume carries on from where they stopped, and the clock does not run while the call is paused. This works the same in every audio experience.
If the character does not answer, the status line says “The character isn’t responding”, and Cesura recovers the call on its own: it asks the character again, and if needed starts a fresh connection that picks up from the conversation so far. The learner keeps their microphone and does not have to repeat themselves. A pause of several seconds before the character carries on is this recovery at work, not a fault.
If the learner’s microphone disconnects during the call, for example a phone used as a wireless microphone, the call switches to the device’s default microphone and carries on. Connecting or disconnecting a headset mid-call does not silence the character.
Twenty minutes, and no continue
Every voice run is capped at 20 minutes. Two minutes out, the learner sees “Time is almost up — this conversation ends in…”. At the cap the audio stops, a Time is up screen appears with no controls, and the run moves itself into the feedback report. There is no way to add time, and closing and reopening the tab does not restart the clock.
If the activity sets a shorter Max Duration, that shorter limit is what the countdown shows and what ends the run.
When the conversation ends on its own
A voice run ends for the same reasons a text one does, and it now says so instead of carrying on with the microphone open:
- The character wrapped things up. The learner sees Simulation Ended and can choose Keep Talking to carry on, or finish and go to the report. The microphone and the character are both muted while they decide, so nothing is said into a conversation that may be over.
- A hard limit was reached — the activity’s message allowance, its duration limit, or the 20-minute cap. These offer no Keep Talking: the run goes to the feedback report.
Ending it yourself
End asks for confirmation and then finishes the run into the feedback report, the same as ending a text simulation. There is no way to leave a voice run half-finished and come back to it.
If the audio will not start, or the call stops
When a call cannot start, the learner sees Audio could not start. Almost always the cause is microphone permission: the browser has to be allowed to use the microphone for the site, and on a phone the page has to be open in a supported browser.
When a call that was under way cannot continue, the learner sees The voice call stopped, with the reason. For example, the microphone disconnected and no other microphone was available, or the connection kept dropping.
Both dialogs offer Try again and End Simulation, and show an Error code: to quote to support. Everything else — the briefing, the forms, the report — works regardless. See Voice and video simulations: what goes wrong and how to fix it.
On a phone
The experience fills the screen on a small device instead of sitting in a panel, with the same controls at the same size. Voice runs work on a phone; a stable connection matters more than it does for text.
On an iPhone, switching to another app can mute the microphone or the sound. When the learner comes back, Cesura tries to restore both. If it cannot, a banner says “We can’t hear your microphone. Tap the screen to reconnect it.” or “The sound is off. Tap the screen to turn it back on.” One tap anywhere on the screen brings it back.
One call per run
A run holds one call at a time, across tabs and windows. If the learner opens the same run in another tab, the call continues there, and the first tab shows The voice call stopped: “This conversation was opened in another tab or window, so it stopped here. Continue there, or choose Try again to bring it back to this page.”