This project demonstrates a complete workflow for building an Emotion Classification Dataset from audio speech files using Label Studio. The goal is to prepare high-quality labeled data for future ...

Issue: The UX flow for creating and leaving annotations on audio files is unintuitive. Process: Currently, the user needs to long press on the waveform, create a default-sized annotation box, press ...

Abstract: There are many sound classification problems that have target classes which are rare or unique to the context of the problem. For these problems, existing data sets are not sufficient and we ...