AI-Powered Audiobook Creation: A Comprehensive Guide to Using Projects

Projects is a powerful feature designed for creating long-form content with AI-generated voiceovers. This innovative tool allows users to transform books, documents, and even entire webpages into professionally narrated audiobooks. With Projects, content creators can streamline their workflow, from text input to final audio output, all within a single platform.

To begin using Projects, navigate to the Projects tab in the menu. Here, you can create new projects or manage existing ones. The number of concurrent projects depends on your subscription level. To create a new project, click 'Create new project' and choose from three options: start an empty project, import an existing EPUB, PDF, or TXT file, or import content directly from a website URL. Each option offers flexibility in how you initiate your audiobook creation process.

When setting up a new project, you'll have the opportunity to customize various aspects. Name your project, select a default voice, choose the AI model, and set quality preferences. It's important to note that while voice settings can be adjusted later, the model and quality settings are locked after project creation. You can also configure download settings, such as audio normalization and metadata inclusion, to ensure your final product meets industry standards.

The Projects interface resembles a familiar online text editor, making it easy to write and edit your content. For imported files or URLs, the system automatically populates the pages and, if well-structured, divides the content into chapters. The editor includes features like undo/redo, insert divider, and heading options to help organize your text effectively. Proper grammar, paragraph structure, and appropriate line breaks are crucial for optimal AI narration results.

Once your content is ready, you can generate voiceovers for specific sections or entire chapters. Select a paragraph (block) and click the play button to generate and play the audio. The system allows for continuous play, enabling you to listen to your audiobook without pre-rendering all paragraphs. You can also regenerate specific sections if you're not satisfied with the initial result. The 'Block Settings' panel allows you to customize voice settings for individual blocks or apply global changes.

To finalize your audiobook, use the 'Convert' button to process either the entire project or selected chapters. After conversion, you can access different versions of your audiobook and choose from various download options. These include downloading individual chapters, the entire project as a single audio file, or a zip file containing separate audio files for each chapter. This flexibility allows you to tailor the final product to your specific needs or distribution requirements.

For precise control over word pronunciation, Projects offers a Pronunciation Dictionary feature. This allows you to upload a lexicon file in .PLS format, specifying how certain words, character names, or acronyms should be pronounced. The dictionary supports both phonetic alphabet (IPA or CMU Arpabet) and word substitutions. This feature ensures consistency and accuracy in the AI's pronunciation throughout your audiobook, particularly useful for unique terms or brand names.