# Le Scribe Audio

> Two-pass AI audio & video transcription to 100% editable Word (.docx)

- URL: https://shipx.live/products/le-scribe-audio
- Website: https://lescribeaudio.com/en
- Maker: @lescribeaudio (https://shipx.live/makers/lescribeaudio)
- Launch week: 2026-W45
- Launched: 2026-11-02
- Categories: AI, Developer Tools
- Tags: transcription, speech-to-text, subtitles
- Pricing: Freemium
- Platforms: Android
- Built with: Next.js, React, Tailwind CSS, Clerk
- For: Content Creators, Enterprises, Small Businesses
- Community: 1 upvotes, 0 comments

## About

Le Scribe Audio is a web-based transcription platform that converts audio and video recordings into text. A two-pass multimodal AI processes raw audio directly, identifies speakers, adds punctuation, and supports more than 100 languages, including switching between two languages in the same file.

After transcription, an interactive player syncs the text with the audio so you can click a paragraph to hear that moment, rename speakers in one click, and export to Microsoft Word (.docx), PDF, or subtitle formats (.srt, .vtt).

Files are stored in France on OVHcloud with AES-256 encryption, inference runs on Google Cloud Vertex AI in Europe with zero data retention, and transcripts can be deleted in one click. A paid human proofreading option is also offered.

It is aimed at professionals who need reliable verbatim records, such as researchers, lawyers, physicians, journalists, and corporate teams.

## The problem

Manually transcribing interviews, meetings, hearings, and lectures takes hours, and standard speech-to-text tools produce errors, lose speaker attribution, and struggle with long or multilingual recordings.

## The solution

Le Scribe Audio sends raw audio to multimodal AI that separates speakers, punctuates and structures the text across 100+ languages, lets you verify passages against synchronized playback, and exports editable Word, PDF, or subtitle files.

## What makes it different

The page emphasizes two-pass multimodal transcription that avoids chaining speech-to-text with post-processing, and combines it with France-hosted, AES-256 encrypted storage, zero data retention, and native support for Olympus and Philips dictaphone files.

## Who it's for

Professionals who need accurate verbatim transcripts of long recordings, such as researchers, lawyers, physicians, journalists, and corporate teams.

---

Launched on ShipX (https://shipx.live), the weekly launch platform for indie makers. The maker wrote this listing; ShipX doesn't verify product claims.