Skip to content

Recording Sessions & Uploading Audio

InstaRecap needs session audio to generate transcripts and summaries. You can upload pre-recorded files, record directly in the browser, or have your AV team upload recordings per session.

Sessions & Transcripts page showing a list of event sessions with upload audio and view transcript actions

Go to InstaRecap → Sessions & Transcripts.

This page shows all your event sessions with their transcription status. You can filter by session type, speaker, or show only transcribed sessions.


  1. Go to InstaRecap → Sessions & Transcripts.
  2. Find the session you want to transcribe.
  3. Click Upload Audio on the session row.
  4. Choose your upload method:
    • Upload File — select an audio file from your computer
    • Record — record audio directly in the browser
  5. The file uploads and enters the processing pipeline.

Standard audio formats are supported including MP3, WAV, M4A, and AAC. The system handles format conversion automatically.


After upload, audio goes through these stages:

StageStatusWhat’s Happening
1QueuedFile uploaded, waiting in processing queue
2ConvertingAudio format being standardized
3ConvertedAudio ready for transcription
4TranscribingAI converting speech to text
5SummarizingAI generating takeaways, summaries, social posts
6ProcessedComplete — transcript and summaries ready

A progress bar shows the current stage. Processing time depends on audio length — a 1-hour session typically takes 5–10 minutes to fully process.


Each recording must be linked to a session:

  1. When uploading, select the target session from the dropdown.
  2. If a recording was uploaded without a session mapping, click Remap to assign it.
  3. One session can have multiple recordings (e.g., if the session was split across two audio files).

For real-time transcription during a session:

  1. Go to InstaRecap → Sessions & Transcripts.
  2. Click Live Transcription on the session row.
  3. Select the transcription language.
  4. Run a sound check to verify audio input.
  5. Click Start to begin real-time transcription.
  6. The transcript appears in real-time as the speaker talks.
  7. Click Stop when the session ends.

Live transcription uses WebSocket connections for real-time audio streaming and supports multiple transcription services.


The sessions list supports:

  • Search — find sessions by title
  • Filter by session type — keynotes, workshops, panels, etc.
  • Filter by speaker — find sessions by a specific speaker
  • Show transcribed only — toggle to see only sessions with completed transcripts
  • Group by day — organize sessions by event day for easier navigation

  • Record everything — even sessions you don’t think will be popular. Attendees often want recaps of sessions they missed.
  • Use lapel microphones — clip-on mics produce much cleaner audio than room microphones. Cleaner audio = more accurate transcripts.
  • Upload during the event — don’t wait until after the event. Upload recordings as soon as sessions end so transcripts are available the same day.
  • Have a backup — use both live transcription and a recording. If live transcription has issues, you can re-process from the recording.
  • Check audio quality — play back a few seconds of each recording before uploading. If you can’t understand the speaker, the AI won’t either.