Every step from audio to accessible text.
A single platform that captures, transcribes, subtitles and translates, with the editor tooling to ship results you can actually publish.
Transcription
File-based or realtime transcription in 48 languages with speaker diarisation and timestamps. Built for legal, government and corporate records.
Subtitling
Generate compliant subtitles for VOD, WebTV and live broadcast. Export to SRT, VTT, EBU-STL and more.
Machine translation
Translate transcripts and subtitles across 30+ languages while keeping timing intact.
The workbench around the engine.
Speech recognition is only half the job. Hover any tile to see what each module does in production.
Realtime processing
Live speech-to-text over WebSocket with licensed realtime entitlement.
Hover for detailsRealtime processing
Supports language selection, live partial and final results, and on-premise realtime model packs when enabled. Designed for broadcast, council chambers and live captioning workflows.
Batch processing
Asynchronous file transcription for uploads or media URLs.
Hover for detailsBatch processing
Language selection, punctuation, diarization, speaker count hints, multichannel, folders, webhooks, templates and caption or transcript document creation, all in one pipeline.
Translation
Document and caption translation into many target languages.
Hover for detailsTranslation
Powered by configured DeepL or Google Translate backends with glossary support. On-premise is disabled by default and can be enabled with third party translation keys. Local vLLM translation is available as a configuration direction.
Custom models
Train organization-specific speech recognition models.
Hover for detailsCustom models
Train on audio and transcript datasets for domain terms, names, acronyms and specialized vocabulary. Upload datasets, validate them, run training, then select successful models during transcription.
Glossaries
Terminology control for translation.
Hover for detailsGlossaries
Create source and target language glossaries from CSV term pairs, keep them active or inactive, and apply them automatically or by selecting a glossary during translation.
Forced alignment
Sync an existing transcript to audio or video timing.
Hover for detailsForced alignment
Upload a transcript with media and produce timestamped transcript or caption output. Useful for preparing training data or turning legacy transcripts into timed subtitles.
Realtime editor
Live transcription workspace for streams.
Hover for detailsRealtime editor
Start and stop streams, choose language and engine, view partial and final text, edit final segments, and monitor connection and stats while broadcasting.
Transcript editor
Editable transcript workspace with media playback.
Hover for detailsTranscript editor
Speaker labels, timestamps, search, split, merge, add and delete utterance controls, save and sync state, translation split view, AI summary, exports, sharing and finalization.
Subtitle editor
Subtitle-focused editor with timed segments.
Hover for detailsSubtitle editor
Media preview, caption search, segment timing controls, time shift, caption styling, split translation view and exports including subtitle formats and burned-in video.
Meeting minutes
Template-based exports and AI summarization.
Hover for detailsMeeting minutes
Admins create export or summarization templates with DOCX or example files and prompts. Users generate meeting-minute-style documents from transcripts in the background and receive or download results.
Speaker directory
Organization-level reusable speaker database.
Hover for detailsSpeaker directory
Store name, role, party, email and notes. Search, sort, create, edit, delete and bulk upload. Editors apply consistent speaker labels across every transcription.
Developer API
REST APIs and webhooks for your workflow.
Open docsDeveloper API
Embed Scriptix into your existing systems. Full technical reference, schemas and examples live at docs.scriptix.io.
Visit docs.scriptix.ioWant the full technical reference?
Dive into our developer documentation.