LibriVox
LibriVox is a volunteer-driven project founded in 2005 that produces free audiobooks of works in the US public domain in many languages. Recordings are dedicated to the public domain in the United States and hosted by the Internet Archive, requiring no account, subscription, or payment to listen or download. Copyright status may differ in other countries.
Score generated by AI agents based on publicly cited evidence and reviewed by the project maintainer. Not independently validated. Last assessed 2026-10-30. Score revised 2026-10-30: 5 → 2.
Score History
Timeline events are AI-curated from public reporting. Score trajectory is derived from documented events.
Hugh McGuire started LibriVox as a blog project on August 10, 2005, and twelve readers signed up for Conrad's The Secret Agent, completed that September alongside a new forum. Proof-listening was formalized in January 2006, Project Gutenberg began listing LibriVox recordings that October, and McGuire explained the choice of a public-domain dedication. The project was small, free, ad-free and founder-led, with informal governance its only mild concern.
The 1,000th audiobook in October 2007, with about 1,500 contributing volunteers and 60-70 releases a month, marked LibriVox's shift from experiment to established library. Growth strained volunteer-built software and paid hosting: the first fundraising drive in February 2010 raised $23,000 from 433 donors in 13 days. In May 2011 a hacker used a superadmin forum account to reach user emails, private messages and encrypted passwords, which LibriVox disclosed and remediated.
In April 2012 LibriVox announced 100 million downloads and an Andrew W. Mellon Foundation grant to rebuild its infrastructure; it moved its core systems to free Internet Archive hosting that November and launched the rebuilt website and catalog in September 2013, followed by a $50,000 donation drive. The catalog reached 10,000 projects in 2016, and LibriSpeech (2015) made the recordings a speech-research benchmark. Setbacks were minor: a temporary 2016 catalog reset and listener confusion over third-party apps' ads.
On January 1, 2019, new works entered the US public domain for the first time in over 20 years, renewing each year the supply of texts LibriVox can record. A 2020 volunteer surge sped production to 15,000 audiobooks by February 2021 and past 20,000 by December 2024, and by April 2023 the project had stopped taking donations for itself. AI-era pressures followed (the Libri-Light dataset, a 2023 human-made content rule, bot traffic behind a 2026 API cap), along with the October 2024 Internet Archive cyberattack, while the core service stayed free, ad-free and account-free.
Alternatives
Free public domain audiobook and ebook site (formerly Books Should Be Free) advertising 7,000+ titles. It draws on public domain sources including LibriVox recordings and Project Gutenberg ebooks, with direct MP3 downloads. It offers another browsing interface for much of the same public domain material.
Free library audiobook app that connects to your public library card, providing access to audiobooks beyond the public domain at no cost. Requires a library card; titles can have wait lists when all copies are borrowed. The available catalog and borrowing limits depend on your library.
Dimensional Breakdown
Summaries below were written by AI agents based on the cited evidence. They are editorial interpretations, not independent research findings.
Dimension History
Timeline (43 events)
Hugh McGuire Launches LibriVox Blog Project
Hugh McGuire created LibriVox on August 10, 2005, on Blogsome.com. The next day, Joseph Conrad's The Secret Agent was selected as its first book and fully subscribed by twelve readers. The project drew inspiration from free software, Creative Commons, Project Gutenberg, and the Internet Archive. LibriVox's About page names Richard Stallman and the free software movement, and Brewster Kahle's talk on universal access to human knowledge, among its influences.
Boing Boing Features LibriVox to Wider Audience
Boing Boing linked to LibriVox, describing it as a podcast of volunteers reading public domain books aloud. LibriVox reported that the attention also put it on the del.icio.us popular page and generated 10,463 hits, and said it planned to call for technical volunteers to help make the site more robust. Reason's April 2007 reporting described Boing Boing as popular and identified early volunteer Kara Shallenberg as having discovered LibriVox through it. Reason also reported that people in the free culture movement had embraced LibriVox.
First Audiobook Completed and LibriVox Forums Launched
All audio files for The Secret Agent were completed and made available, making it LibriVox's first finished audiobook. The project simultaneously launched its web forum at librivox.org with 6 books already in progress, establishing the collaborative infrastructure that would drive the project's growth for the next two decades.
Internet Archive Invites LibriVox to Open Library Launch
The Internet Archive invited LibriVox to participate in the official launch of the Open Library, a project affiliated with the Open Content Alliance. LibriVox provided a recording of Henry James' An International Episode, which Brewster Kahle played during his demonstration. The report proposed a future formal partnership to record more Open Library books.
Proof-Listening Forum Opens for Quality Control
LibriVox opened a dedicated Proof-Listening forum, formalizing its quality control process. Volunteer proof-listeners would review recordings for technical issues like repeated passages, long silences, editing errors, and volume problems before cataloging. The process followed published guidelines but deliberately excluded subjective judgments about reading style or accent.
First Dramatic Audio-Play Completed and Cataloged
The Importance of Being Earnest by Oscar Wilde became LibriVox's first completed audio-play, expanding the project beyond straight readings into dramatic performances with multiple voice actors. This demonstrated the community's ability to coordinate complex multi-reader projects and broadened the types of public domain works the project could make accessible.
LibriVox Recordings Added to Project Gutenberg Catalog
Project Gutenberg began listing LibriVox audiobook recordings on its catalog pages, formalizing the symbiotic relationship between the two public domain projects. LibriVox sourced texts from Project Gutenberg, and Gutenberg in turn directed users to LibriVox's audio versions. This partnership expanded the reach of both projects without any commercial arrangement.
Creative Commons Features LibriVox Public Domain Dedication
Creative Commons published a feature on LibriVox in which founder Hugh McGuire explained the project's decision to dedicate recordings to the public domain. He said the team considered various Creative Commons licenses but chose the Public Domain Dedication because it best fit LibriVox's spirit of unrestricted access.
Reason Magazine Profiles LibriVox as Model of Peer Production
Reason Magazine published 'The Wealth of LibriVox,' a feature article framing the project within Yochai Benkler's theories of peer production. The article included comments from Jimmy Wales and David Joseph, Audible.com's vice president for marketing, positioning LibriVox alongside Wikipedia as a leading example of volunteer-driven internet projects that produce genuine public goods without commercial incentives.
LibriVox Reaches 1,000 Completed Audiobooks
LibriVox published its 1,000th audiobook, Edgar Allan Poe's The Murders in the Rue Morgue, read by Reynard T. Fox. At this point the project was releasing 60-70 books per month with approximately 1,500 volunteers who had contributed audio, with recordings in 21 languages. LibriVox was already the most prolific audiobook publisher in the world by volume.
LibriVox Reaches 1,500 Audiobooks at 70+ Per Month
LibriVox reached 1,500 completed audiobooks, releasing over 70 titles per month with 115 released in May 2008 alone. Creative Commons described the growing catalog as a collaborative culture project releasing recordings into the public domain.
LibriVox Publishes 2,000th Audiobook
LibriVox released its 2,000th audiobook, Edward Gibbon's The History of the Decline and Fall of the Roman Empire Vol. VI, reaching this milestone just 14 months after passing 1,000. The announcement thanked readers, coordinators, proof-listeners, moderators, and technical volunteers, as well as the Internet Archive and Project Gutenberg. The international contributor base was documented earlier: in a December 2006 Creative Commons interview, McGuire described an influx of Italian and German volunteers and volunteer-led website translations.
LibriVox Reaches 3,000 Audiobooks
LibriVox published its 3,000th audiobook, The Red Planet by William John Locke, on December 26, 2009. By February 4, 2010, the catalog totaled two full years of continuous audio (17,520 hours), demonstrating the output of its volunteer-driven operation.
LibriVox Launches First Community Fundraising Drive
LibriVox published 'LibriVox Needs Your Help,' its first direct fundraising appeal, seeking $20,000 to cover hosting costs (~$5,000/year) and improve website usability. The project was handling approximately 1,800 GB of Internet transfer per month.
LibriVox Reports $23,000 Raised in 13-Day Fundraising Drive
LibriVox closed its first fundraising campaign after raising $23,000 from 433 donors in 13 days, exceeding its $20,000 target. It thanked the Internet Archive for managing the donations and said it would not have to talk about money again for a few years.
LibriVox Celebrates Fifth Anniversary
Founder Hugh McGuire marked LibriVox's fifth anniversary by recalling its original volunteer audiobook experiment. The full anniversary post reported 3,889 readers and an average of 87 books per month over the previous six months, and thanked readers, proof-listeners, coordinators, coders, and system administrators.
LibriVox Describes Existing Catalog API and OPDS Development
LibriVox described its existing public catalog API, which had already enabled third-party apps and web services. Chris Goringe was working on an OPDS catalog to standardize developer access; the announcement suggested it might become the first audiobook catalog using that specification.
LibriVox Forum Hacked, Database Exposed
A hacker compromised a superadmin account on the LibriVox forum, accessing the complete database including email addresses of all users, private messages, and encrypted passwords. The attacker vandalized forum pages and injected HTML into templates. LibriVox responded by disabling the compromised account, cleaning up vandalism, restricting admin database access, and forcing all users to reset passwords.
LibriVox Reaches 5,000 Completed Audiobooks
LibriVox welcomed its 5,000th project, Roderick Hudson by Henry James, read by Nicholas Clifford. The milestone was reached four years after the 1,000th audiobook, adding another four thousand projects over that interval. Solo recordings now comprised a significant portion of the catalog alongside collaborative projects.
LibriVox Reaches 100 Million Downloads, Receives Mellon Foundation Grant
LibriVox announced 100 million downloads of its free audiobooks from the Internet Archive, coinciding with an Andrew W. Mellon Foundation grant to rebuild technical infrastructure and improve website accessibility. Internet Archive director Brewster Kahle called LibriVox's collection 'one of the most popular' on the Archive.
LibriVox Moves Core Systems to Internet Archive Hosting
LibriVox moved its website, forum and other systems from servers hosted by Syntenic to a new home at the Internet Archive, taking the site and forum offline from November 28 to November 30, 2012. Audiobooks stayed available at archive.org during the move, though the LibriVox catalog could not be used to find them.
LibriVox Invites App Developers to Test New API
LibriVox invited developers to test its revamped catalog API and directed them to a forum thread containing the latest information and documentation for building third-party apps and web services.
LibriVox Launches $50,000 Infrastructure Fundraising Drive
LibriVox sought $50,000 by its eighth anniversary for infrastructure beyond the Mellon grant's scope, ongoing software development, internationalization, and big-ticket costs volunteers would otherwise pay themselves. The appeal said day-to-day audiobook work ran with no paid staff, that infrastructure hosting at the Internet Archive was now free, and that a donation page would remain afterwards.
Mellon-Funded Website Redesign and Catalog Rebuild Launches
LibriVox launched its completely rebuilt website and catalog system, funded by the Mellon Foundation grant. The redesign made browsing and discovering audiobooks significantly easier, while the underlying volunteer workflow software was rebuilt from scratch. At the time, the catalog contained over 7,000 free audiobooks, making LibriVox 'certainly the largest publisher of free public domain audiobooks.'
LibriSpeech ASR Corpus Derived from LibriVox Published
Researchers Vassil Panayotov, Guoguo Chen, Daniel Povey, and Sanjeev Khudanpur published the LibriSpeech corpus at ICASSP 2015, containing 1,000 hours of read English speech derived from LibriVox audiobook recordings. The dataset became one of the most widely used benchmarks in automatic speech recognition research, demonstrating an unintended but valuable secondary use of LibriVox's public domain recordings.
LibriVox Reaches 10,000 Completed Projects
LibriVox published its 10,000th audiobook, including 5,556 solo recordings and 1,349 projects in 36 non-English languages. The milestone demonstrated the project's scale as the world's largest publisher of free public domain audiobooks. Production had been sustained at roughly 1,000 new titles per year for five consecutive years.
Server Failure Resets LibriVox Catalog
A temporary server failure reset LibriVox's catalog to its state as of September 15, 2016, causing recently cataloged books to disappear from the homepage. All audiobooks remained safely stored on the Internet Archive, and users could access and download them directly. LibriVox volunteers worked to restore the catalog, with the incident highlighting the importance of the Internet Archive as redundant storage.
LibriVox Addresses Third-Party App Ad Confusion
LibriVox clarified that it has 'nothing to do with' third-party app development and no say in the apps' content. Listeners had complained about political ads embedded in recordings and attributed those viewpoints to LibriVox; some said they would stop recommending or volunteering for the project. LibriVox explained that public-domain status permits commercial use and directed listeners to its free, ad-free originals.
Public Domain Day Returns After 20-Year Freeze
For the first time in over 20 years, published works whose copyrights remained in force began entering the US public domain again on January 1, 2019. Copyright terms on works published in 1923 had been extended to 95 years by the 1998 Copyright Term Extension Act. The newly available works included 1923 publications by Robert Frost, P.G. Wodehouse, and Edith Wharton, expanding the public-domain material available for projects such as LibriVox to record.
Digital Humanities Quarterly Documents LibriVox Volunteer Labor
Digital Humanities Quarterly published 'The In/Visible, In/Audible Labor of Digitizing the Public Domain,' a peer-reviewed study examining how LibriVox volunteers perform the labor of curators, copyright researchers, digital content managers, voice artists, project managers, mentors, translators, and audio producers. The study highlighted that much of this labor remains invisible to end users who download finished audiobooks.
Facebook AI Releases Libri-Light with 60K Hours of LibriVox Audio
Facebook AI Research released Libri-Light, a dataset containing 60,000 hours of unlabelled speech extracted from LibriVox audiobook recordings with audio from 7,439 unique speakers. The dataset was designed for training automatic speech recognition systems with limited supervision. Published at ICASSP 2020, it demonstrated the continued use of LibriVox's public domain recordings as foundational training data for AI speech technology.
LibriVox Reaches 14,000 Audiobooks
LibriVox celebrated its 14,000th audiobook with Lullaby-Land by Eugene Field, reaching the milestone in 11 months. The catalog contained 1,813 projects in 41 languages other than English, with more than 9,800 readers, proof-listeners, and coordinators contributing.
LibriVox Celebrates 15,000 Audiobooks in Record Time
LibriVox reached 15,000 audiobooks nine months after the previous milestone. It said it welcomed 'an amazing amount of new volunteers' in 2020 as one of the 'few positive side-effects of the Covid19 pandemic.' The project thanked more than 10,850 readers, including over 1,000 new readers since the previous milestone in June 2020.
LibriVox Reaches 16,000 Audiobooks
LibriVox celebrated its 16,000th audiobook eight months after reaching 15,000, maintaining the faster catalog growth reported during the pandemic. It thanked more than 11,700 volunteers and reported 2,044 audiobooks in languages other than English.
LibriVox Marks 17 Years with 17,000 Audiobooks
LibriVox celebrated a symmetrical milestone: 17 years of operation with 17,000 completed audiobooks. The project maintained its accelerated production rate, now averaging approximately 1,200-1,500 new titles per year compared to roughly 1,000 per year before the pandemic volunteer surge.
LibriVox Extends Human-Made Content Policy to Summaries and Covers
In 2023, LibriVox admins extended its existing human-made-recordings policy to project summaries and cover images. Detected AI-generated content is removed; repeated violations may lead to warnings and a permanent ban. The policy says LibriVox has not sold and will not sell recordings, while public-domain status allows AI trainers to use them without permission or payment. The policy explicitly says the prohibition protects LibriVox's reputation, integrity, and community culture of personal contributions by human volunteers.
LibriVox Stops Accepting Donations for Itself
LibriVox's donation page, citing 17,898 completed works as of April 1, 2023, says the project will no longer accept donations specifically for LibriVox until a new need arises and asks supporters to donate to the Internet Archive, which hosts its forum, catalog and software systems. The date is approximate: the page dates its catalog figure, not the policy change.
LibriVox Reaches 18,000 Audiobooks in 9 Months
LibriVox celebrated its 18,000th audiobook after completing another thousand books in nine months. The project reported more than 13,000 volunteer readers, roughly 1,000 more than nine months earlier, and recordings in 100 languages, including 2,260 stand-alone works in 46 languages besides English.
Internet Archive Cyberattack Disrupts LibriVox Audio Host
An October cyberattack on the Internet Archive exposed information from more than 31 million user accounts, including email addresses and encrypted passwords, and forced a temporary shutdown of its services. LibriVox identifies the Internet Archive as the host of its audio files, making Archive availability an infrastructure dependency for its downloads.
LibriVox Catalog Reaches 20,000 Audiobooks
LibriVox's catalog reached 20,000 recordings on December 17, 2024, roughly doubling since its 10,000-project milestone in August 2016. Its 19th-anniversary post earlier in 2024 reported 19,584 projects in 48 languages and more than 13,000 readers. The August 2025 anniversary later reported 14,276 volunteers and described LibriVox as one of the world's largest producers of free audiobooks.
Internet Archive Publishes Public Domain Spotlight on LibriVox
The Internet Archive blog published a 'Public Domain Spotlight' feature on LibriVox, celebrating the project's 20-year history and its role as 'a crucial resource, ensuring that our cultural heritage is freely and openly accessible.' The feature highlighted that all recordings cost nothing, have no limitations on listening time, and are devoid of DRM.
LibriVox Celebrates 20th Anniversary
LibriVox marked its 20th anniversary with 20,648 completed audiobook titles and 14,276 volunteers. The anniversary post described the project as entirely volunteer-run and looked ahead to another twenty years. Its community also created an anniversary collection of works with 'twenty', 'twentieth', or '20' in their titles, cataloged on August 4, 2025.
LibriVox Announces 500-Record Cap on Catalog API Requests
Noting a catalog of more than 21,000 completed projects, LibriVox said traffic from bots, scrapers and some apps had become a problem for the volunteers who share its server. It set a maximum of 500 records per API request from a target date of September 26, 2026, said the entire catalog could still be retrieved in batches with the offset parameter, and said it might later rate-limit individual clients.
Evidence (54 citations)
D1: User Value Erosion
D2: Business Customer Exploitation
D3: Shareholder Extraction
D4: Lock-in & Switching Costs
D5: Twiddling & Algorithmic Opacity
D6: Dark Patterns
D7: Advertising & Monetization Pressure
D8: Competitive Conduct
D9: Labor & Governance
D10: Regulatory & Legal Posture
Scoring Log (6 entries)
5→2. D5 1→0 (correction: the fact audit corrected the 'proof-listening introduces procedural opacity' basis; proof-listening follows published technical guidelines, and the 2026 API change was pre-announced with rationale), D8 1→0 (correction: the old score rested on third-party apps' commercial conduct, which the fact audit placed outside the core project; the Sept 2026 500-record API cap keeps the whole catalog accessible), D10 1→0 (recalibration: no lobbying, litigation or enforcement; proactive published copyright policy; the only blemish is the 2011 forum breach, now scored in its own era). Eras: Blog Origins re-dated 2005-08-01→2005-08-10 (founding); Volunteer Scaling re-dated 2008-01-01→2007-10-31 (1,000th audiobook); Institutional Growth re-dated 2013-01-01→2012-04-05 (Mellon grant/100M downloads) and relabeled Mellon-Funded Rebuild; 20K Catalog Maturity re-dated 2026-02-17→2019-01-01 (US Public Domain Day ends 20-year freeze) and relabeled Public Domain Thaw. Era scores 1/3/2/2. Since Oct 2025: only material change is the Sept 2026 API request cap (500 records, bot/scraper load) alongside a catalog of 21,000+ projects; no ads, fees, legal or governance changes found. Historical gap-fills: 2010 fundraising outcome, Nov 2012 move to Internet Archive hosting, July 2013 $50,000 drive, ~April 2023 end of donations for LibriVox itself (dated from the page's 'as of April 1, 2023' figure; Wayback copies rate-limited). Timeline[15] (API/OPDS 2011) impacts set to 0 per handoff. Alternatives unchanged. Category 'Audiobooks' fits.
Checked 16 trimmed candidates: 0 restored, 4 partly restored, 10 confirmed removed, 2 already present. PARTLY RESTORED timeline:update:0: named Stallman/Kahle influences, LibriVox About; timeline:update:1: Boing Boing recruited Kara Shallenberg and was popular, plus free-culture support, Reason 2007-04-24; timeline:update:12: international volunteer influx/translation work moved to documented 2006 context, Creative Commons 2006-12-05; timeline:update:34: human-community protection rationale, published LibriVox AI policy. Added evidence: PARTLY RESTORED founding influences (LibriVox About), recruitment (Reason), and international volunteer growth (Creative Commons). Existing primary milestone citations remain, supplemented by these claim-specific evidence entries. ALREADY PRESENT timeline:update:13: December 26 event and February 4, 2010 17,520 hours; evidence:update:27: actual volunteer roles on original guide. CONFIRMED REMOVED timeline:update:11: unchanged quality/proof-listening keeping pace; timeline:update:15: anniversary claims of organic growth/self-sustaining governance; timeline:update:30: onboarding without governance degradation; timeline:update:31: pandemic-cohort retention; timeline:update:33: permanent cohort growth/non-reverting output; timeline:update:35: continuing uploader, separate surviving store and multi-week unavailability (SN_BlackMeta/Zendesk already have evidence); timeline:update:36: >14,000 readers at December 2024 and by-far-largest ranking (dated 2024/2025 snapshots already retained); timeline:update:38: absence of mission drift/pressures; evidence:update:2: app problems/date incorrectly attributed to the 2019 article; evidence:update:22: alternatives list proves healthy competition. All original pages were read live, including cached full source text. Wayback was attempted for the 14 candidates requiring further research; no readable archived copy resulted, and failed requests were not evidence. Used 28 search queries in 7 WebSearch calls, no more than 3 queries per candidate. See work/restore-decisions.json for dropped subclaims, verdicts, search counts and supporting URLs.
Audited 80 timeline/evidence/milestone items plus 2 alternatives and all prose. Supplied Team Mode ACCEPTs count as verified; all48FIX and1REVIEW decisions resolved. Eleven supplemental evidence citations retain already-present facts on their actual sources. Corrected role attribution: timeline[8] calls the Audible interviewee CEO, while Reason identifies David Joseph, vice president for marketing (https://reason.com/2007/04/24/the-wealth-of-librivox/). No specific figure, named person, study, or verbatim quote was established as invented under the strict definition; incorrect role attribution, unsupported claims and date placeholders are reported as corrections, not fabrication. Preserved true facts almost trimmed by verifiers: founding emails, McGuire's2008 CC argument, 17,520h February2010, fundraising result/partner donations, paid Mellon contractors, LibriSpeech benchmark uptake, PD2019 authors, loud app ads, and20th collection. Removed two duplicates. See work/lead-decisions.json and work/prose-findings.md for complete decisions.