Skip to main content

Welcome to Vox101

Vox101 is a real-time transcription and translation platform for live events. Speak into a microphone and the platform produces:

  • a transcript in the source language,
  • one or more live translations into target languages, and
  • optional spoken translation (text-to-speech) for spectators.

Audio sessions can be recorded, stored, and summarized after the fact.

How this documentation is organised​

  • User Guide — the place to start. A guided introduction that walks you through running your first event, from preparing a vocabulary to downloading the recording afterwards.
  • Administration — for users with the Admin role only: inviting users, changing roles, resetting passwords, and managing the organisation's device fleet.
  • Everything else — standalone reference pages, such as Your profile, that you can dip into whenever you need them.

Core concepts at a glance​

  • Organisation — your workspace in Vox101. All events, vocabularies, recordings, and devices belong to an organisation. A single user account can belong to multiple organisations.
  • Event — a named gathering (conference, lecture, meeting, ...) scheduled in time, with one or more source and target languages. Live sessions and recordings are produced inside an event.
  • Session — a single transcription run for an event. Sessions capture audio, produce transcripts and translations, and optionally store a recording.
  • Vocabulary — a piece of supporting material that improves recognition accuracy: free-text background, glossary terms, translation pairs, language hints, and document attachments.
  • Device — a physical or browser endpoint that participates in a session as a source (microphone) or display (subtitles, mirror, chromakey).
  • Spectator — anyone watching a live event via a public link; they receive captions and optional spoken translations without signing in.