Stenovia is a software platform for producing live transcripts of parliamentary sessions, court hearings, conferences, and similar events. It is built around the workflow of working stenographers and scopists, and combines three capture sources — machine stenography, AI speech recognition, and manual editing — in one shared editor.
Stenographers write natively on their steno machine into the Stenovia editor. At any point during a session, the operator can press a freely configurable stroke to insert text from the live AI speech-recognition stream into the document. Steno and speech-recognition output sit on the same line; there is no separate window or buffer.
The capture mode can be changed at any time during the session:
Switching modes happens via a single stroke and does not interrupt the session.
Multiple editors can open the same document at the same time. Cursors, selections, and edits are visible to all participants in real time. Every change is attributed to a user and stored in the document history.
The platform includes:
In addition to live AI capture, Stenovia is developing a set of AI tools for post-session editing of the transcript:
Automatic correction suggestions: the AI scans the document for likely transcription errors, terminology inconsistencies, and formatting issues, and proposes corrections inline. The editor accepts, rejects, or modifies each suggestion individually.
Command-driven rewrites: the editor can select a passage and issue a natural-language command — for example, to standardise speaker labels, reformat a list, or convert between direct and indirect speech. The AI proposes the change as a diff that can be applied or discarded.
Bulk operations: apply terminology, style, or formatting rules across the entire document in a single pass, with full version-history tracking of every AI-generated edit.
All AI operations are transparent: every change is logged with the prompt that produced it, attributed as AI in the audit trail, and fully reversible.
Each word in the transcript is anchored to a timestamp in the audio recording. Clicking a word in the editor plays from that exact position. Foot-pedal integration is supported for playback control (rewind, slow, advance, jump-to-word).
The platform includes a complete live audio pipeline: capture, low-latency streaming, native synchronisation between audio and text, and live distribution to viewers.
The client-facing side is separate from the working editor:
Live notes can be sent to the client during the session.
Release editor: operators select which segments are visible to the client. Unreleased material stays inside the team workspace.
Clean viewer: the client view does not show timestamps, scopist annotations, or work-in-progress markers.
Live listen-in: clients can connect to the live audio stream and follow the released viewer in parallel.
Round-trip corrections: client feedback can be re-imported into the editor and merged with the working version.
Stenovia runs natively on macOS and Windows. The client viewer runs in any modern browser and requires no installation.
The platform is designed for distributed teams. Stenographers, scopists, editors, and clients can be in different locations and work on the same session in real time.
Stenovia is developed by SD Steno Deutschland GmbH, Berlin.
Website: https://stenovia.com/ Email: hello@steno-deutschland.de