Voyce leans on video as much as phone, which means your camera framing is part of the job, not just your voice. Here's how to run a live transcription and translation layer beside that work — privately, on your side of the screen only.
A tablet rolls up to a hospital bed, a nurse angles it so the patient and the camera both fit, and you're mid-diagnosis with four seconds of context. Voyce built its practice around exactly that moment — video remote interpreting for healthcare, with phone interpreting alongside it for when a screen isn't practical or bandwidth won't hold one. On video, your framing and your face are as visible as your voice, which changes what the job actually demands of you. This page is a practical guide to running a live caption and translation layer beside that work, written for the interpreter.
Voyce's interpreting network is built specifically around healthcare access, with video remote interpreting as its most visible service — a tablet, a mobile device, or a browser window connecting a provider and a patient with limited English proficiency, on demand, usually in seconds. Phone interpreting runs alongside it for the calls where video isn't practical: low bandwidth, a phone-only intake line, a patient who can't manage a screen. Either way, the calls land the same way — no chart, no visit summary, no warning about what's coming.
Phone interpreting asks you to hold everything in your ears and your memory. Video interpreting adds a channel: you're also tracking a screen — the patient's expression, the provider's body language, your own framing, sometimes a chart held up to the camera — while the audio keeps moving at the same pace it always did. That split attention is a real tax, and it sits on top of the standard load of rendering medical language accurately in real time.
The effect compounds across a shift the same way it does in any high-volume remote interpreting work — quietly, then all at once. We go into the mechanics in interpreter cognitive load and burnout; the practical takeaway here is that anything that moves detail off your working memory and onto a screen you control is capacity you get back for the next call.
Unicaption is a real-time captioning and translation copilot that runs on your own device, next to the tablet, laptop, or platform you're taking Voyce calls on. It listens to your microphone or your system audio only — it never joins the video feed, never appears on the patient's or provider's screen, and never touches the call itself:
For a wider look at copilots built specifically for interpreters, see our honest comparison with Interpreter.
Set your audio source once — system audio if calls play through your computer's speakers or a browser tab, microphone if they come through a headset or a separate device. Position the caption window off to the side, out of your camera's frame. Skim your dictionary.
Keep your eyes where they'd normally go — the patient, the provider, the screen — and glance at the transcript the way you'd glance at notes on paper. It's there to catch what you didn't, not to be read from. Our guide to telehealth live captioning and translation covers video-specific setup.
The same session runs whether the call is video or audio-only — no need to reconfigure between a VRI assignment and an OPI one back to back.
Add whatever term slowed you down today to the dictionary before you forget which patient it came from. It costs a minute now and saves one on every future call.
Healthcare interpreting runs on protected health information whether the call is on video or audio only, and being visible on a video call doesn't change what confidentiality demands of anything running beside it. Unicaption's design reflects that: sessions are end-to-end encrypted, audio is never stored, transcripts are deleted the moment the session ends, and nothing is ever used to train models. There's no recording of a video call sitting somewhere afterward, because Unicaption was never in the video call to begin with — it only ever had your microphone or system audio.
What we won't do is tell you Voyce's own policy on aids and tools during a session — that's theirs to set and yours to know. Check before you assume, the same way you would with any resource you bring to a call.
That's a question for Voyce's own policies, which you should check directly rather than assume. What we can tell you is what Unicaption does: it runs locally on your device, reads only your microphone or system audio, never joins the video feed or the call itself, never records, and deletes transcripts when the session ends. Whether an aid like this is appropriate on a given assignment is a professional judgment that stays with you and your agency's guidance.
No. Unicaption isn't a participant in the video session, a bot, or an overlay on the shared screen. It captions your microphone or your computer's system audio entirely on your own device — the patient and the provider see the same call they'd see without it, because there's nothing added to see.
Yes. Because Unicaption reads audio rather than integrating with any specific video or phone platform, the same session works whether you're on a video remote interpreting call or a phone-only one — nothing to switch or reconfigure between the two.
Every account includes 30 free minutes each week, renewing automatically, no credit card required. Paid plans start at $9/month (Instant) and go up to $149/month (Ultra), with annual discounts. Try it across a few real shifts on the free minutes before deciding whether — and which plan — makes sense.
30 free minutes every week, no credit card. Nothing joins the call, nothing appears on screen.
Start Free Trial →