Voice Interviews
How browser-based AI voice interviews work, from the participant's perspective.
How It Works
Voice interviews run entirely in the participant's browser — no app downloads or plugins required. An AI interviewer speaks questions aloud using a natural-sounding voice, listens to the participant's spoken responses, and generates follow-up probes in real time.
The experience feels like a natural conversation, not a robotic Q&A. The AI adapts its follow-up questions based on what the participant says, probing deeper into interesting responses and gracefully moving on from brief ones.
Participant Experience
Welcome screen
Participants see your welcome message, study description, and optionally enter their name and email. They agree to proceed before starting.
Onboarding
A brief 3-slide carousel explains how the interview works — how to respond, what to expect, and tips for a good conversation.
Device setup
Participants grant microphone access (and camera for video mode). They can select their preferred microphone and test that audio is working.
Connecting
A brief connecting screen while the AI agent initializes. Usually takes 1-3 seconds.
Interview
The main conversation. The AI asks questions, the participant speaks their answers, and the AI follows up. A progress bar shows how far through the interview they are.
Thank you
A completion screen thanking the participant. They can be redirected to a custom URL if configured.
Pause and Resume
Participants can pause a voice interview at any time using the pause button. Their progress is saved and they can resume within 7 days.
- Paused sessions save all conversation history and progress
- The resume link is stored in the participant's browser (localStorage)
- On resume, the AI picks up exactly where it left off with full context
- Paused time is not counted against voice minutes
- After 7 days, the session expires and cannot be resumed
Video Mode
Video mode adds a camera feed on top of the voice interview. The participant's video is recorded alongside the conversation for analysis. The AI still speaks and listens via audio — the video is a recording layer, not a video call.
Keyboard Shortcuts
During voice interviews, participants can use the spacebar to toggle their microphone on and off. This provides a push-to-talk option that some participants find more natural than clicking a button.