LibriVox
LibriVox coordinates volunteers around the world who record public domain books as audiobooks and release them for anyone to use. Because the source texts are out of copyright and the recordings are dedicated to the public domain, you can use the audio freely, including for training and commercial work.
For RAG and machine learning, the value is in pairing spoken audio with known source text. LibriVox recordings underpin widely used speech datasets such as LibriSpeech, and they are a practical starting point for speech-to-text training data or for multimodal systems that need to retrieve across both audio and text.
Recordings vary in audio quality and reader accent because they come from many different volunteers, so plan for some filtering if you need consistent audio.
Related sources
Google Books Ngrams
Word and phrase frequency data drawn from Google's digitised book collection, spanning centuries of published text. It counts how often words and short phrases appear by year, which is useful for historical language analysis and time-aware RAG features.
Open Library
An Internet Archive project building a web page for every book ever published. It holds more than 20 million catalogue records and lends many titles digitally, making it a rich source of book metadata for RAG systems.
Project Gutenberg
A volunteer effort to digitise and archive public domain books, with more than 70,000 free ebooks. Mostly English, but it covers many languages. One of the oldest digital library projects, so it is a clean, permissively licensed source of full-text literature for RAG.
Standardized Project Gutenberg Corpus (SPGC)
A research-ready version of the Project Gutenberg catalogue with consistent formatting, tidy metadata, and token counts for every book. It saves you the cleanup work, so you get uniform full-text literature ready to chunk and embed for RAG.