About Le Scribe Audio
Le Scribe Audio (published by LE SCRIBE AUDIO, France) is an AI audio and video transcription web platform built for CSE/CSSCT secretaries, public administrations, legal professionals, researchers, and journalists. Its two-pass transcription engine with speaker diarization cuts drafting time in half (2 to 3 hours of full re-listening and formatting per audio hour instead of 4 to 7 hours of manual typing from scratch). Users review transcripts in a synchronized interactive audio player, globally rename speaker labels (diarization), generate an on-demand executive summary followed by a chrono-thematic breakdown, and export a standard 100% editable Word (.docx) file, PDF, TXT, or SRT/VTT subtitles. Files are encrypted in France on OVHcloud (AES-256) with 1-click deletion, AI inference runs via Google Cloud Vertex AI (Europe region, Zero Data Retention), and optional human proofreading by two successive transcribers is available at 60 EUR excl. VAT/hour.
Le Scribe Audio is a French B2B web platform dedicated to professional AI audio and video transcription, published by LE SCRIBE AUDIO (EURL, Le Rheu, France). Designed for CSE and CSSCT committee secretaries, public sector institutions, town halls, legal practitioners, academic researchers, and journalists, Le Scribe Audio streamlines transcript preparation while maintaining rigorous editorial and GDPR standards. Using a two-pass AI transcription pipeline with automatic speaker diarization and timestamps, the platform cuts drafting and review time in half: a full human re-listening, correction, and formatting pass takes 2 to 3 hours per hour of recorded audio, compared to 4 to 7 hours of manual typing from scratch. Inside the browser-based synchronized interactive player, users click any word to jump to the exact audio moment, globally rename or code speaker labels (such as replacing generic speaker tags with Enquete P1 or M. Martin across all turns), and generate an on-demand AI summary structured into an executive summary followed by a chrono-thematic breakdown. Completed transcripts export as a standard 100% editable Microsoft Word (.docx) document, PDF, plain text (.txt), or SRT and VTT subtitle files. Audio files and transcripts are stored and encrypted in France on OVHcloud (AES-256) with immediate 1-click permanent deletion, while AI inference operates via Google Cloud Vertex AI (Europe region) with contractual Zero Data Retention (no training or retention on customer files). New users receive a 60-minute free trial without a credit card, lifetime hour packs start at 7.50 EUR excl. VAT, and an optional human proofreading service performed by two successive transcribers is available at 60 EUR excl. VAT per recorded hour.
