User Guide

Welcome to Whombat!

We're happy to have you on board with Whombat, the open-source tool for audio annotation with ML in mind. Whether you're just getting started or looking to refine your annotation process, this guide is here to make your work smooth and enjoyable.

Here we you can get more information about how to start your audio annotation work and make best use of this tool.

Table of Contents

1. Installation
2. Registering Recordings and Managing Datasets
3. Navigating the Spectrogram
4. Creating Annotation Projects
5. Visualizing and Evaluating Model Predictions
6. Exploration of Data
7. Import Example Data
8. Export Annotations for ML Development

Where to start

Here's how to get started:

1. Install Whombat: Follow the step-by-step instructions in the installation guide to set up Whombat on your machine.
2. Explore Whombat's Features: Familiarize yourself with Whombat's core functionality by reviewing the user guides for each main section:
   • Datasets: Learn how to manage and organize your audio data and associated metadata.
   • Annotation Projects: Discover how to create and manage annotation projects, define tasks, and collaborate with others.
   • Evaluation Sets: Understand how to create standardized evaluation sets for assessing model performance.
   • Exploration: Explore the tools and visualizations for interactive data exploration and analysis.
3. See Whombat in Action: For a hands-on experience, import the example data provided with Whombat. This will allow you to explore the interface and functionality with a pre-populated dataset and annotation project. Follow the import data guide for detailed instructions.

Searching for Answers

Feel free to use the handy search bar above to quickly locate the information you need. Additionally, we've curated a FAQ section to address common queries and provide even more insights.

Can't find what you're looking for? No worries! Reach out to us with your questions.