[HN Gopher] LibriVox
___________________________________________________________________
LibriVox
Author : bookofjoe
Score : 243 points
Date : 2025-06-01 21:02 UTC (1 days ago)
(HTM) web link (librivox.org)
(TXT) w3m dump (librivox.org)
| Moosdijk wrote:
| I wonder if AI will be a benefit or a detriment to this project.
|
| On the one hand, there's going to be a lot more, potentially high
| quality audio books in its repository, on the other hand it goes
| against the spirit of the project itself.
| woodson wrote:
| The speech data collected by this project has been used for
| more than a decade to build automatic speech recognition and
| text-to-speech synthesis systems (see LibriSpeech, LibriTTS,
| LJSpeech). It definitely has been a benefit to AI.
| joshstrange wrote:
| I think they are talking more about the impact of AI on
| Librivox, as in people running an ebook through an AI TTS
| tool and uploading it.
|
| On one hand, a well curated/edited AI recording might be
| great but a lot of people will (try? Idk their policies) to
| upload AI slop (no proof-listen, no checking, just laziness).
| woodson wrote:
| I think that, for the purposes of creating high-quality
| Free audiobooks, the issues are essentially the same with
| human-generated recordings as with AI generated ones. The
| recording quality and faithfulness to the original text
| (both in terms of "content" and the appropriate reading in
| terms of tone, expression of emotion, etc.) have to be
| verified. The problem is scale. There will be many more
| TTS-generated recordings uploaded than human-generated
| ones. Some automated filters (e.g., ASR WER, audio quality
| metrics) would be a great first step to discard bad-quality
| slop right away (though it might unfairly penalize real
| human accented speech).
|
| Importantly, the recording should indicate whether it was
| human or AI generated.
| username223 wrote:
| > Importantly, the recording should indicate whether it
| was human or AI generated.
|
| This is all that's necessary. Sometimes I'm fine with
| mediocre TTS; sometimes I want an actual professional;
| librivox is somewhere in between, but should clearly
| specify whether I will be getting an amateur human or a
| robot.
| swores wrote:
| I disagree, for the reasons stated by the person you
| replied to.
|
| Historically, being told that a voice recording is AI
| generated would be enough to tell you to expect basic TTS
| robotic voice, but with advances in AI voice generation
| we're approaching the point where AI can sound as good as
| real humans - it's not yet to the point where it's easy
| to generate an audiobook as good as a professional
| reader, but that point will come in the not too distant
| future.
|
| And equally on the other side, something being recorded
| by a human doesn't automatically mean it has the quality
| of a professionally-read audiobook. This is something
| LibriVox has always had to deal with, by gatekeeping
| which volunteer recordings to either give feedback
| requesting improvements to or to not use at all.
|
| In some but not all cases, an amateur human reader can
| already be as good as a professional, that will soon be
| true for AI. For both AI and humans it will remain the
| case that some efforts are not as good, but the line
| between them (for quality) isn't going to be whether or
| not they are AI - though I do agree that AI or not should
| also be labelled.
| username223 wrote:
| Certainly TTS has improved a lot thanks to modern AI, but
| it simply doesn't have the information to improve beyond
| sounding like a human reading words fluently. A
| professional audiobook reader modulates his tone to
| reflect narrative mood, chooses voices for the characters
| consistent with their natures, etc., and transformer
| models can't do those things.
|
| For an example of a professional audiobook, check out Rob
| Inglis' version of The Lord of the Rings.
| swores wrote:
| I agree with you about the current stage of things, which
| is why I said that we're approaching the point but not
| yet there for AI to be able to match professional
| readers.
|
| But I disagree with you when you write "it simply doesn't
| have the information to improve beyond sounding like a
| human reading words fluently" - it has the same
| information when reading it as a human does, meaning that
| the best implementation would have to not only adapt tone
| to explicit instructions like "... she shouted", but also
| read between the lines / make subjective choices to suit
| the different characters.
|
| AI is already capable of doing sentiment analysis on
| text, and text to speech models are getting better at
| being able to simulate moods/emotions rather than just
| speaking flatly, and I don't think we're many years away,
| if that, from those two sides being paired together in a
| way that produces the sort of quality output we're
| talking about for the first time without human
| involvement. Add to that the fact that AI can train on
| the many good examples of humans reading things, they may
| get to the point of emulating not just the core accent
| but also how each accent should adopt to what meanings in
| the text and arrive at a great solution without even
| needing to go through the steps of analysing what the
| text means to use that to know how to modify the voice
| being generated.
| username223 wrote:
| You're more optimistic about this stuff than I am, but I
| think I get your perspective. We have decent sentiment
| analysis, fluent text generation, and real-sounding TTS,
| so combining them will yield a pretty good reading. I
| agree that you're probably right when it comes to
| newspaper columns and magazine articles, but that's not
| on the level of a good audiobook.
|
| To take an example, here's an iconic line from the
| Fellowship of the Ring:
|
| > The wizard swayed on the bridge, stepped back a pace,
| and then again stood still. 'You cannot pass!' he said.
|
| If you think that is a command, you should shout it like
| Ian McKellen in the movie. If you think it's a statement
| based on superior knowledge (see
| https://acoup.blog/2025/04/25/collections-how-gandalf-
| proved...), you should probably state it with certainty
| and fatigue. And if you're making a movie with a ton of
| crazy special effects and swelling music, you should
| probably make whatever choice goes best in that context.
|
| Even if a model could make some consistent choice there,
| I wouldn't be all that interested, because the reader
| conveying their interpretation of the character to the
| listener is what matters. Sure, it might get enough
| Spotify plays to make some money, but it's not art.
| reddit_clone wrote:
| I get very annoyed at the AI voice overs in youtube
| shorts and videos (which are showing up more and more
| lately).
|
| I just close the tab when I realize it is AI. Not sure
| how long I can do this.
| kristopolous wrote:
| I imagine there's various disabilities where audio readings
| greatly simplify people's lives. They're probably
| appreciative of anything accurate regardless of whether
| it's humans talking or not.
| UmYeahNo wrote:
| Well, you can safely assume that everything in Librivox was
| used to train the AI. So, "benefit" or "detriment"... you make
| the call.
| spudlyo wrote:
| I'd like to think that in the future this will be possible, but
| at present I believe there are still too many uncanny valley
| problems for me to regard any TTS generated audio books as high
| quality. I can sometimes tolerate listening to articles or
| technical essays done this way, but quality audio book
| narrators often do consistent and distinct character voices,
| understand complex emotional states, and are capable of
| reacting to contextual clues and subtext. This seems like a
| pretty high bar for current TTS models.
|
| Something like NotebookLLM seems shockingly good at first, and
| gives me hope that eventually we'll have machines that are
| nearly as good as humans at this; but after listening to it for
| an hour or so the novelty wore off and the artifice of it now
| seems galling and distracting.
| bookofjoe wrote:
| >quality audio book narrators often do consistent and
| distinct character voices, understand complex emotional
| states, and are capable of reacting to contextual clues and
| subtext.
|
| The best experience I've ever had with audio books is John le
| Carre reading his early novels (not in public domain). He
| uses a different voice for each character and they are SO
| pulsing with life it's breathtaking.
| spudlyo wrote:
| Thanks for the tip! I love well narrated audiobooks, and
| I've been meaning to get into one of le Carre's books for a
| while. I see he narrates the "John le Carre Value
| Collection" on Audible which has Tailor of Panama, Our
| Game, and Night Manager for a single credit. Is that what
| you're referring to?
|
| Along those lines, there is a great 2007 unabridged
| audiobook[0] of Frank Herbert's Dune that is read by Simon
| Vance for narration, but other characters are dramatically
| performed by other voice actors. It's excellent, but sadly
| a tad bit uneven and inconsistent in production. It's like
| they got 3/4ths of the way through the project and some of
| the original voice actors couldn't complete the project and
| Vance had to pick up the slack. Regardless, it's still one
| of my favorite audiobooks.
|
| [0]: https://libro.fm/audiobooks/9781427201447-dune
| bookofjoe wrote:
| My favorites are his first books read by him:
|
| "Call For The Dead" and "A Small Town in Germany"
|
| Listen here:
|
| https://youtu.be/e1lmpG3kCDg
|
| https://youtu.be/30QOqAcY4bY
|
| https://youtu.be/0Ik9Gv9s0TQ
|
| https://youtu.be/q79SspzdpLA
|
| https://youtu.be/i3UnPBMouwU
|
| FWIW The three novels in the "Value Collection" are
| abridged:
|
| https://www.amazon.com/John-le-Carr-Value-Collection-
| audiobo...
| SubGenius wrote:
| I've enjoyed quite a few very well narrated audiobooks on
| LibriVox. The Jane Austen novels voiced by Karen Savage are
| phenomenal.
| towledev wrote:
| Haven't listened to Librivox in years and years, but I still
| fork over the annual $2.99 because I feel I owe it.
|
| It's horizon-broadening. Lots and lots of interesting
| reads/listens I never would've picked up otherwise. 1800s ghost
| stories, darkly racist novels like The Leopard's Spots (good
| luck getting through the first 10 pages). My favorite is
| Havelock the Dane: A Tale of Old Grimsby, first written circa
| the 14th century but thought to be much older. When you listen
| to it, it is apparent that the author and the intended audience
| know 100x more about nautical things than you do. It's also
| charmingly simplistic; the main character is sort of like Conan
| the Barbarian. He'll do things like "lift a stone the weight of
| an ox and throw it the length of two men." You imagine the
| audience being like, "Oh my fucking god.... that's amazing."
| Nopoint2 wrote:
| Elizabeth Klett is good too.
| dlcarrier wrote:
| I once couldn't find the audiobook of a book that my book club
| was reading, and it was a long book at I didn't have time to set
| aside to solely reading. It turned out that there weren't any
| commercially produced audiobooks of it, but it was public domain,
| and I found it on LibriVox.
|
| The book was long and boring, but at least the narrating was
| good.
| waltbosz wrote:
| https://librivox.org/the-adventures-of-sherlock-holmes-versi...
|
| David Clarke does a very good narration of many Sherlock Holmes
| stories.
| begemotz wrote:
| "This is a libravox recording. All libravox recordings are in the
| public domain..."
|
| I contributed to a Robert Lynd project on libravox many years
| ago, and I still remember saying these intros.
|
| (Lynd was a wonderful essayist, if anyone might be interested).
| jfengel wrote:
| I know th books are always in the public domain. Are the
| recordings public domain as well, or are they under a free
| license like GPL?
| harrisi wrote:
| All LibriVox recordings are in the public domain.
|
| https://librivox.org/pages/about-librivox/
| WillAdams wrote:
| One of my favourite novels, H. Beam Piper's _Little Fuzzy_ is in
| the public domain (because he died intestate and the Commonwealth
| of Pa. failed to renew his copyrights before they were turned
| over to his estate).
|
| There is a professional-quality reading of it by Tabithat:
|
| https://librivox.org/little-fuzzy-by-h-beam-piper/
|
| which I recommend highly.
|
| Unfortunately, the quality of the readings can vary widely, and
| my family has often been unwilling to put up with a poor quality
| recording on long trips.
|
| That said, I use their app on my phone while doing boring tasks
| at work, and greatly appreciate the project.
| rectangleboy wrote:
| > Unfortunately, the quality of the readings can vary widely .
| . . I tried LibriVox in college to listen to difficult-to-
| understand poems (at the professor's recommendation) on my iPod
| and my ears couldn't take the harsh plosives that nearly every
| recording I tried had.
| WillAdams wrote:
| There is a collection of different authors reading
| Coleridge's "Kublai Khan" which is a great way to get an
| experience of a lot of different readers and possibly find
| one you like.
|
| For my part, I'm grateful that folks volunteer their time and
| energy, and when there is only one reader for a given text,
| accept it.
| Der_Einzige wrote:
| It really is the e-book version of trying to do "netflix n
| chill" with FOSS.
|
| It's sad that the best tech projects are not useful to normies.
| nl wrote:
| I don't think "netflix n chill" means what you think it does.
| linsomniac wrote:
| If you're one of today's lucky 10,000: It is slang for
| having sex, beyond the literal meaning.
| Der_Einzige wrote:
| It means what I think it means.
|
| Do you think a girl wants to deal with you trying to get
| your torrent streaming system working with your TV? Hell
| no!
| kps wrote:
| If she doesn't, it wouldn't have worked out anyway.
| abawany wrote:
| FWIW they now accept donations via the Internet Archive:
| https://wiki.librivox.org/index.php?title=Donate .
| photochemsyn wrote:
| A real gem is the Adrian Praetzellis reading of Kim by Rudyard
| Kipling:
|
| https://librivox.org/kim-by-rudyard-kipling/
| theschmed wrote:
| Another site, which includes a smaller but more professionally
| curated set of recordings, is Lit2Go
| (https://etc.usf.edu/lit2go/books/). My children and I for
| example have greatly enjoyed Lorraine Montgomery's recording of
| "Curly and Floppy Twistytail", a series of delightful nonsense
| stories performed with gusto.
| (https://etc.usf.edu/lit2go/183/curly-and-floppy-twistytail-t...)
| They're not all aimed at children either, high quality recordings
| of Dracula, David Copperfield, etc
| Min0taur wrote:
| Ohhh, I never thought I'd see FCIT/lit2go mentioned on HN! I
| used to work in the sound booth for lit2Go, it was one of my
| favorite jobs. We had some wonderful vocal talent come through,
| and I learned much of what I know about recording/mixing from
| that work.The people in charge of organizing that project were
| diligent/earnest to the point of sainthood.
|
| They have quite a few historic stereo-views scattered around
| their open source pic site, I always enjoyed browsing those
| with my 3d shades on:
|
| https://etc.usf.edu/clippix/search?q=stereoview (They had more
| than this search would suggest; hopefully they're still
| around...)
| Tryk wrote:
| Elisabeth Klett is phenonmenal, way beyond any professional
| recording I've purchased.
|
| https://librivox.us/search.jsp?search=reader%3A%22Elizabeth+...
| miki123211 wrote:
| I wonder if we could make LibriVox, but with Eleven Labs.
|
| Sure, Eleven Labs isn't anywhere near as good as professional
| narration (yet), but it's better than your average Librivox
| reader for sure.
|
| At this point, it's just a matter of throwing money at the
| problem (and not that much money at that), and that's usually
| easier than finding talent.
| parkersweb wrote:
| I think Eleven Labs has essentially done that already with
| ElevenReader haven't they?
| Deprogrammer9 wrote:
| https://librivox.org/anarchy-by-errico-malatesta/
|
| this is the way.
| spookie wrote:
| I really appreciate the existence of this project and all people
| who have contributed to it.
|
| I want to contribute a reading of a book someday in my native
| tongue, as it is slowly dying (less than 1k speakers). A way of
| preserving it.
| rendall wrote:
| What is it?
| spookie wrote:
| I fear it would be too personally identifiable lmao
|
| It's within the EU though
| tinco wrote:
| Your name implies it might be in The Netherlands? Perhaps
| Bildts?
| bArray wrote:
| I think there needs to be a way to apply multiple filters, i.e. I
| want Science Fiction _and_ to listen in English. It would be good
| to also be able to combine with keywords too.
___________________________________________________________________
(page generated 2025-06-02 23:01 UTC)