Analyze the Full Scope of Legal Evidence, Not Just the Documents
By Steven Van ·
Harvey now supports audio and video transcription along with enhanced image understanding.
Harvey now transcribes audio and video files and can visually interpret images, charts, and diagrams, rather than reading only the text around them. Previously, evidence like exhibit images or deposition recordings had to be reviewed by hand or sent to a separate transcription vendor before Harvey could work with it.
- Image understanding: upload a standalone image or a PDF, Word, or PowerPoint file with visual content, and Harvey analyzes the relevant page to answer questions, useful for ownership structures in purchase agreements or patent drawings.
- Audio transcription: drop a file into Assistant or upload it to Vault for a speaker-labeled, timestamped transcript. Supports M4A, MP3, WAV, WebM, FLAC, and OGG, up to 500MB in Assistant or 4GB in Vault, and up to two hours per file. The Harvey mobile app can also record depositions or client meetings directly, saving the transcript to Vault with speakers labeled; only the transcript is retained, not the original audio.
- Video transcription: the same process works for MP4, MOV, AVI, DAV, and WebM files, up to two hours each, letting litigation teams bulk-upload deposition videos into Vault and query across the set, including surfacing contradictions or pulling references into a Review Table.
Harvey announced the update on 14 August 2026.

Harvey
AI built for legal work — agents, document analysis, and legal research trusted by 75+ AmLaw 100 firms and in-house teams in 70+ countries.
View Harvey →