Capture and analyze user research sessions with one API
Research platforms capture insight that lives or dies in the recording. MeetStream joins research sessions on Zoom, Google Meet, or Microsoft Teams and returns recordings, per-participant video, and diarized transcripts built for analysis.
Meet participants where they already are
Recruiting research participants is hard enough without asking them to install a new tool. Moderated sessions happen on whatever platform the participant or the research team prefers, and a research platform that only supports one of them loses sessions at the margins.
MeetStream puts a recording bot into Zoom, Google Meet, and Microsoft Teams sessions through the same API. No host permission or paid meeting plan is required, so it works for sessions scheduled by researchers, by participants, or by a recruiting agency in the middle.
Per-participant video makes highlight reels easy
The most persuasive research artifact is a clip of a real user reacting. MeetStream provides separate video and audio files per participant, so you can cut highlight reels from the participant's stream alone, without the moderator's face or voice bleeding into every clip.
The full-session MP4 is there for context, and per-participant audio means quotes pulled for reports are clean and correctly attributed even when the moderator and participant talk over each other.
Transcripts researchers can actually tag
Diarized transcripts separate moderator prompts from participant responses, which is the structure every downstream research workflow depends on: tagging, theming, quote extraction, and search. Timestamps map every utterance back to its moment in the video, so a tagged insight is one click away from the footage that supports it.
Transcripts are available async after the session, and in real time via webhook if your product offers live notes during moderation. If a session's audio was difficult, you can re-trigger transcription on the stored recording.
One request to join any session
Instrumenting a session is a POST request with the meeting URL, and a webhook tells your platform when the recording and transcript are ready to process. For scheduled studies, calendar integration with Google and Outlook can place bots on every upcoming session automatically, up to three months ahead.
From session to searchable repository
Because each session arrives as structured pieces, feeding a research repository is straightforward: embed and index the diarized transcript, attach clips from per-participant video, and file it all under the study using meeting metadata like participants and duration. Sessions become searchable evidence instead of videos nobody rewatches.
Participant data is treated with care throughout: encryption in transit and at rest, GDPR compliance, ISO 27001 certification, and bring-your-own S3 when research media needs to stay in your own storage.
How it works
- Send a POST with the session's meeting URL, or let calendar integration schedule bots for the study.
- The bot records the session while your moderator focuses on the participant.
- Receive the MP4, per-participant video and audio, and a diarized transcript when the session ends.
- Pipe everything into your repository for tagging, theming, and highlight clips.
Frequently asked questions
Does this work for both moderated and unmoderated research?
It covers any session that happens as a live meeting on Zoom, Google Meet, or Microsoft Teams, which is where most moderated research runs. Unmoderated tests recorded inside your own product would use your existing capture path.
What format do recordings come in?
You get an MP4 of the full session plus separate audio and video files for each participant, along with metadata like who attended and when they joined and left. That combination covers both playback and clip-cutting workflows.
How quickly are transcripts ready?
Real-time transcript segments with speaker labels can stream to your webhook during the session, and the full async transcript is available after it ends. Re-triggering transcription is available whenever you want a fresh pass.
How do we try it before committing?
Sign up at app.meetstream.ai, grab an API key, and use the included free credit to record a few test sessions. The docs at docs.meetstream.ai walk through the first bot in a few minutes.
Build it on MeetStream
Spin up your first meeting bot in minutes. One API to join, record, stream, and transcribe across Zoom, Google Meet, and Microsoft Teams.