Image-based approaches to classifying audio data

This invention uses artificial intelligence to determine if spoken audio is real or fake. It works by converting sound into a visual representation, like a sound wave graph, and then a computer program trained on examples of real and fake speech analyzes that picture to decide if new audio is authentic. The system specifically focuses on speech and can even check small parts of a conversation.

Why it matters: Deep learning models for image and audio processing have advanced significantly since 2018, making robust authenticity detection more feasible and necessary given the rise of synthetic audio.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedHeavy software system, machine learning model training

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10