Public-domain books from Project Gutenberg and open-access research papers from arXiv, Semantic Scholar and OpenAlex, under Explore in the app. Add one to your library and press play.
Search by title or authorLanguageEnglish
EN
Books
FictionI
Science
History
Philosophy
Poetry
AdventureII
Research
Computer ScienceIII
Medicine
Physics
Biology
Mathematics
Psychology
Chemistry
Economics
Only the file is downloaded.
ExploreBooks · English
Project Gutenberg · No. 84
Frankenstein
or, The Modern PrometheusMary Wollstonecraft Shelley1797–1851
“You will rejoice to hear that no disaster has accompanied the commencement of an enterprise which you have regarded with such evil forebodings.”
Add to library
EPUB
English
Gothic fiction
Science fiction
IFiction
Pride and PrejudiceJane AustenPG 1342Jane EyreCharlotte BrontëPG 1260Great ExpectationsCharles DickensPG 1400Moby-DickHerman MelvillePG 2701Wuthering HeightsEmily BrontëPG 768Crime and PunishmentFyodor DostoevskyPG 2554A Tale of Two CitiesCharles DickensPG 98The Great GatsbyF. Scott FitzgeraldPG 64317Anna KareninaLeo TolstoyPG 1399The Picture of Dorian GrayOscar WildePG 174DraculaBram StokerPG 345Dr Jekyll and Mr HydeRobert Louis StevensonPG 43
IIAdventure
Treasure IslandRobert Louis StevensonIn the readerThe Adventures of Sherlock HolmesArthur Conan DoylePG 1661Alice’s Adventures in WonderlandLewis CarrollPG 11Peter PanJ. M. BarriePG 16The Time MachineH. G. WellsPG 35
arXiv:1706.03762 · 2017Attention Is All You NeedVaswani et al.AbstractThe dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration.Open access · PDFarXiv:1512.03385 · 2015Deep Residual Learning for Image RecognitionHe et al.AbstractDeeper neural networks are more difficult to train.Open access · PDFarXiv:1412.6980 · 2014Adam: A Method for Stochastic OptimizationKingma and BaAbstractWe introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments.Open access · PDFarXiv:2006.11239 · 2020Denoising Diffusion Probabilistic ModelsHo, Jain and AbbeelAbstractWe present high quality image synthesis results using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics.Open access · PDFarXiv:1406.2661 · 2014Generative Adversarial NetworksGoodfellow et al.AbstractWe propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a discriminative model D.Open access · PDFarXiv:1810.04805 · 2018BERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingDevlin et al.AbstractWe introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers.Open access · PDF
In the readerTreasure IslandRobert Louis StevensonVoiceSerenaSpeed1×
Fig. 1 · Library
Shelves you can play.
Explore in the app: books from Project Gutenberg by topic, papers from arXiv, Semantic Scholar and OpenAlex by field. Add one to your library and the reader plays it.
How it works2
Find a shelf, or search.
Under Explore in the app, browse books by topic and papers by field. Search books by title or author, papers by title or topic.
Pick a book or a paper.
Books come from Project Gutenberg as EPUB, papers as their open-access PDF. Add one and it is saved to your library.
Press play.
It opens in the reader like your own files, and each sentence is highlighted as it is read.
Notes3
Two shelves, one reader.
The library has two shelves. One holds books from Project Gutenberg, the volunteer library of public-domain books: novels, poetry, history and philosophy whose copyright has run out. The other holds research papers, found in three open indexes at once: arXiv, Semantic Scholar and OpenAlex. Whatever you pick opens in the same reader as your own files, with the spoken sentence highlighted as it goes.
Books are sorted by topic: fiction, science, history, philosophy, poetry and adventure. You can also search them by title or author. Choose a language and the shelves show books written in it, in any of the 31 languages the reader speaks.
Papers are sorted by field, from computer science and medicine to economics, and you search them by title or topic. Results from the three indexes come back as one list, newest first, with duplicates removed.
Use it for the reading you keep meaning to do: the classic you never finished, played while you cook or walk, or a paper from your field, heard once through before you sit down with the figures.
A few limits are worth knowing. Only open-access papers are listed, and only those with a PDF or a DOI to find one by. When a paper’s own link leads to a publisher’s page instead of the file, SpeakCove looks for a free copy by its DOI; if there is none, the paper can’t be opened. Books arrive as EPUB with their chapters, papers as the authors’ PDF. Picking one downloads that file and nothing else and saves it to your library, so it is there when you are offline.
Books by title or author, papers by title or topic
Language
Books in any of the 31 reader languages; papers have no language filter
Which ones
Public-domain books; open-access papers with a PDF or a DOI
Opens as
Books as EPUB with their chapters, papers as PDF
Privacy
Only the chosen file is downloaded. No account, and no personal data is sent.
Questions5
Are the books really free?
Yes. Project Gutenberg books are in the public domain in the United States, so they are free to read and to listen to. Copyright terms differ in other countries, and Gutenberg asks readers there to check their local law.
Can SpeakCove read research papers aloud?
Yes. In the app, search arXiv, Semantic Scholar and OpenAlex under Explore, pick an open-access paper, and its PDF opens in the reader like any other file. A paper’s PDF you already have also opens in the free reader on this website.
Are there books in languages other than English?
Set the language to any of the 31 the reader speaks and the shelves show Gutenberg books in it. Most of the collection is in English, so some languages have only a few.
Can I listen to a book I already have?
Yes, if it is DRM-free. PDF, EPUB, MOBI, AZW3, DOCX, FB2, CBZ, RTF, TXT and HTML open in the app and in the free reader on this website.
Try it6
Pick a book and press play.
The library is under Explore in the SpeakCove app for iPhone and iPad. The books and papers on its shelves are free.
You can hear Project Gutenberg's public domain books three free ways: AI-narrated audiobooks from the Project Gutenberg Open Audiobook Collection, volunteer recordings on LibriVox, or any of its 75,000+ ebooks read aloud by an AI voice in SpeakCove. Here are 10 of the best books to start with.
The best free audiobook websites are LibriVox (21,000+ public domain books read by volunteers), Libby and hoopla (library audiobooks with a library card) and the Project Gutenberg Open Audiobook Collection. SpeakCove is different: AI voices, not human narrators, read any public domain or DRM-free book aloud, free in your browser.
The best free sci-fi audiobooks are the public domain classics that invented the genre, such as Wells' The Time Machine (about 3.5 hours), Shelley's Frankenstein and Verne's Journey to the Center of the Earth. SpeakCove reads all 10 below aloud with an AI voice for free, and LibriVox has volunteer recordings of many.
The best free self-help audiobooks are the public domain classics modern self-help borrows from: Seneca's Of the Shortness of Life (under an hour), Epictetus' Handbook (about an hour) and Franklin's Autobiography with its 13-virtue plan. SpeakCove reads all seven free picks aloud with an AI voice; three modern titles follow.
AlexNet, ResNet, the Transformer, BERT and six more landmark machine learning papers, with accurate citations, a reading order and free PDFs you can read or listen to.
SpeakCove
