Evernote adds AI-powered transcription for audio, video, and images

Evernote’s new AI Transcribe feature represents a significant value addition to the note-taking platform following its difficult transition under Bending Spoons ownership. The multimodal transcription tool, which quickly and accurately converts audio, video, and image files to text, marks a notable technological advancement for the service that has recently faced criticism for staff reductions and price increases.

The big picture: Evernote has launched a powerful AI-based transcription tool that works across multiple media formats with impressive speed and accuracy.

• The feature is available both within the main Evernote application and as a standalone web tool for easy access.
• Users can upload existing files, record audio directly, or even paste video URLs for instant transcription.

How it works: The AI Transcribe tool processes audio, video, and image files, converting spoken words and written text into editable digital content.

• The system handles files up to 100MB or 60 minutes in length, with special optimization for transcribing Zoom meetings in MP4 and MOV formats.
• For images, the tool can extract text from handwritten notes and whiteboards, supporting both JPG and PNG formats.

Performance details: Testing shows the transcription capability functions efficiently even with challenging content.

• A 55-minute podcast in Brazilian Portuguese was accurately transcribed in under one minute, demonstrating the tool’s processing speed and multilingual capabilities.
• After transcription, users can copy, download, or edit the resulting text directly.

The catch: Evernote’s generous features come with significant usage limitations for non-paying users.

• The free plan restricts users to 50 notes, a single notebook, 200MB maximum note size, and access from only one device.
• Paid subscriptions start at $14.99 monthly or $129.99 annually, positioning the tool primarily for committed users rather than casual transcription needs.