Mato interviews are audio-first. Every interview captures a full audio recording regardless of whether the camera is on. But if your guest turns on their webcam, Mato records video alongside the audio. That video can later be used for social media clips that pair the guest's face with the AI host.
Video is always opt-in. Guests choose whether to enable their camera before joining, and they can toggle it off at any time during the conversation.
Where the camera toggle appears
On the briefing page (the final step before joining the interview room), guests see a camera card near the bottom of the screen. It asks "Turn on camera?" with a toggle switch and a short explanation: video is optional, clips may be used in social posts, and they can change their mind during the interview.
When the guest flips the toggle on, the browser asks for camera permission. If granted, a live preview of their webcam appears below the toggle so they can check framing and lighting before joining.
If the browser denies camera access (or the device has no camera), a short error message appears under the card. The guest can still join the interview with audio only.
How consent works
Consent is recorded server-side before the guest enters the interview room. When a guest enables their camera and clicks Join interview, Mato sends a request to save a consent timestamp on the invite record. The timestamp is set by the server clock, not the browser, so it is an auditable record of when consent was given.
If that consent request fails for any reason (network issue, server error), the guest is redirected into the room with audio only. Mato never records video without a stored consent record.
A timestamp is recorded server-side the moment you enable your camera and click Join interview. This record is permanent and cannot be overwritten or backdated after the fact.
Inside the interview room
When a guest joins with video enabled, the interview room shows their camera feed as the main stage area on the left side of the screen. A small recording indicator ("REC") appears in the top-right corner of the video feed while the interview is active.
The guest can toggle their camera off and on at any point using the camera button in the bottom toolbar of the stage area. Turning the camera off pauses the video track without stopping the audio recording. Turning it back on resumes video capture without needing to re-grant browser permissions.
If the camera was denied or no camera was detected, a small status label appears in the bottom-right corner of the room ("Camera access was denied" or "No camera found"). The interview continues normally with audio.
What happens to the video file
When the interview ends, Mato uploads the video track alongside the two audio tracks (guest and host). The upload sequence:
- The client requests signed upload URLs from the server, specifying that a video track is included.
- Audio tracks are uploaded first (audio is the priority).
- The video track uploads separately. If the video upload fails, the audio recording is still saved. A warning logs the failure, but the interview is not lost.
- After all uploads finish, the client notifies the server with the storage paths.
Video is recorded directly in your browser and stored securely alongside the audio recordings.
The video goes through a status lifecycle tracked on the interview session record: none, recording, uploading, processing, ready, or error. Only sessions with a ready status display the video on the detail page.
Viewing video on the interview detail page
After the interview is complete and the video upload finishes, the interview detail page shows a Video Recording card below the audio recordings section.
The card includes a native HTML video player with standard playback controls (play, pause, scrub, fullscreen). The label reads "Video recording of [guest name]" to make it clear whose camera was captured.
Video only appears when the session's video status is ready. If video was not captured, or if processing failed, the card does not render at all.
Video and social clips
When video is available for a completed interview, the social clips section on the detail page can generate video clips. The Generate clips section checks for video availability and passes that flag to the clip generation pipeline.
Privacy and data handling
A few things to know about how Mato handles interview video:
- Opt-in only. The camera is off by default. Guests must actively enable it and the browser must grant permission before any video is captured.
- Consent is recorded. A server-side timestamp marks exactly when the guest opted in. This record cannot be overwritten.
- Audio is never at risk. Video upload failures do not affect the audio recording. Audio tracks upload first, and video upload is non-blocking.
- Guests can turn the camera off mid-interview. Toggling the camera off pauses the video track immediately. The audio recording continues uninterrupted.
- Access through the dashboard. Video files are stored alongside audio recordings. Only team members with access to the podcast can view the video on the detail page.
- No video is captured without stored consent. If the consent API call fails before joining, the guest enters with audio only, even if their camera was enabled on the briefing page.