Le Scribe Audio is an AI transcription platform that converts audio and video files into text. It processes recordings such as interviews, meetings, lectures, and dictations. The platform uses a two-pass multimodal AI model that analyzes raw audio to generate text transcripts, supporting over 100 languages and code-switching between languages within a single file. The platform includes automatic speaker identification (diarization) and a synchronized audio player that allows users to click on any part of the text to hear the corresponding audio segment. Users can rename speakers across the document with a single click. Transcriptions can be generated in verbatim or polished modes, with an optional AI summary. Exports are available as Microsoft Word (.docx), PDF, and subtitle formats (.srt, .vtt). File storage is handled on OVHcloud servers in France with AES-256 encryption, and AI inference is performed through Google Cloud Vertex AI in Europe under a strict Zero Data Retention policy. Users can delete their data with one click. The platform is used by academic researchers, legal professionals, works council secretaries, and media teams. Users upload files ranging from 10 seconds to 10 hours in length. Supported formats include MP3, WAV, M4A, MP4, MOV, and DS2. The platform processes the audio and presents the transcript in an interactive editor with synchronized playback. A human proofreading service is available for additional verification. The service offers 60 minutes of free transcription upon account creation without requiring a credit card. Subsequent transcription is available starting at €1.50 per hour of audio.