Data labelling

What is data labelling?

Data labeling or data annotation constists in adding tags or labels to raw data such as images, videos, text, and audio. Labels provide context so that a machine learning model can learn from it. For example, labels might indicate whether a photo contains a bird or car, which words were uttered in an audio recording, or what is the main sentiment in text.

What is Eden AI data labelling tool for?

Custom Vision: you will need to labellize your images to perform custom image classification. You will set a or multiple labels to each image of your dataset, then it will generate a file that you will be able to diretly input into custom image classification platform / API on Eden AI. You will also need to labellize your image for Eden AI custom object detection. Here, you will need to draw the bounding boxes around objects and define their label.

Text Processing: Text labellization can be used for multiple features on Eden AI. You can annotate entities (custom NER) and sentiments (custom sentiment analysis) in texts. You can also add a label to each text for custom text classification. Finally, you could have to annotate text data for custom machine translation, in htis case you would have to set the translation of some words, texts.

Audio Processing: Audio processing can require you to first manually transcribe it into written text. You can add tags to categorize the audio. This categorized audio becomes your training dataset for custom speech-to-text.

Eden AI labelling tool allows you to generate your dataset file in the right format for your custom AI projects. But it can also be useful for your benchmarks. You can compare your labels with results of pre-trained AI API to measure performances of different providers.

How Eden AI data labelling tool works?

With Eden AI data labelling tool, you can easily import your data, choose the type of annotation then annotate data per data with a user firendly interface.