Cogniflow AI playground

Try some of our ready-to-use AI models in seconds.

This Web App was created in minutes using our Smart Extractor↗︎ feature.

Want to use this Receipt Extractor or any Image Extractor directly from your iPhone?
Try our iOS shortcut↗︎

Upload an image of a receipt to get the date, store name, VAT number, and total.
👇
Locate elements or objects and count them. Works with 80 categories like person, dog, cat, cars, bottle, etc.
Compare two different photos of a person to verify if it is the same person.
You can use passports or IDs.
Transcribe the first 30 secs of an audio in any language

CogniGPT - Chat with your docs

This is our newest Question Answering system powered by ChatGPT.
In this example the chatbot below has been trained with the content of the new videos released by YC Startup School.
More info here: https://www.ycombinator.com/blog/startup-school-videos

You can also build this chatbot for your documents and embed it on your website!

Identify and count objects in images

This is our Object detection pre-trained model using the COCO dataset which identifies 80 types of objects such as persons, dogs, cats, and many common life objects like tables, books, computers, cell phones, etc.

You can also build your own object detection model using Cogniflow. Learn more in this tutorial

Extract text from images or PDFs

OCR models recognize and extract text from images or scanned documents. They convert printed or handwritten text into editable and searchable formats. This model is typically used in digitizing physical documents, and automating data entry.

Audio transcription

Speech recognition models convert spoken language into written text. They can transcribe audio recordings or interpret real-time speech. Speech recognition is utilized in virtual assistants, voice-controlled systems, transcription services, and accessibility applications.

Face Similarity

Face Recognition models analyze facial features and patterns to determine like in this case, the similarity between two faces. These models are commonly used in applications such as social media platforms or security systems to recognize identity by comparing the face of a person with their ID or passport.

NER (Named Entity Recognition)

NER models extract and classify named entities, such as names, locations, organizations, or dates, from text. These models help analyze and understand unstructured text data and are used in information extraction, text summarization, or search engine applications.