Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Abogen – Generate audiobooks from EPUBs, PDFs and text

github.com

Post reply on HN
Print view View original

31–40 of 85 posts

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#31

Post by hombre_fatal »

rkagerer wrote: post #5
8s2ngy wrote: post #4

I've been using Kokoro TTS with the CLI app, audiblez, mentioned in the "Similar Projects" section of the README. The model is fast and delivers impressive quality for its small size. Some issues I have faced, however, are: a) It doesn't distinguish periods at the end of sentences from the dots in abbreviations such as "Mr." or "Mrs." The result is an awkward pause between "Mr." and the name. b) It doesn't handle ell…

The Mr. / Mrs. thing feels like it would be a pretty easy fix, at least to eliminate a lot of the more common cases.

^ A thought that everyone has had at one point when processing human text before learning the hard way (like end of sentence detection). :P

The difference is that even weak LLMs are good at magically doing this, so I wonder what the problem is for the TTS mentioned above.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#33

Post by leobg »

hombre_fatal wrote: post #31
rkagerer wrote: post #5

Earlier quoted context omitted.

The Mr. / Mrs. thing feels like it would be a pretty easy fix, at least to eliminate a lot of the more common cases.

^ A thought that everyone has had at one point when processing human text before learning the hard way (like end of sentence detection). :P The difference is that even weak LLMs are good at magically doing this, so I wonder what the problem is for the TTS mentioned above.

Kokoro is small and fast because all the text -> phoneme conversion is done by “dumb code” and only the phoneme -> sound part is done using a neural net.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#34

Post by robin_reala »

ekianjo wrote: post #19
floppyd wrote: post #14

I tried Kokoro for voicing blog posts and articles and wasn't impressed to be honest. Right now Gemini 2.5 Flash TTS is a much more capable system with generous free limits (about 10 minutes per generation and about 90 minutes per day). Voices are not very consistent between generations, but for shorter pieces it's not a big deal (but will obviously be for books)

Kokoro is fine for TTS, but it lacks emotion. But for a model of this size, that is kind of given.

Ironic given the name: kokoro is Japanese for heart or sentiment.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#35

Post by frumiousirc »

TOGoS wrote: post #3

The demo video doesn't seem to have any audio in it! At least none that either ffmpeg or whatever Firefox uses can recognize.

Thanks for this. I thought I had some local issue with waterfox. Pasting the (long) video URL to the terminal let's mpv play it with audio.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#36

Post by throwup238 »

hulitu wrote: post #9
nikolayasdf123 wrote: post #2

can I choose any voice? would love to read software engineering books in voice of Morgan Freeman, or maybe even better, Scarlett Johansson

Why not Stephen Hawking ?

Because the Stephen Hawking voice spends a quarter of its time joking/complaining how it never got a Nobel Prize.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#37

Post by pyman »

Thorrez wrote: post #27
JSR_FDED wrote: post #20

Earlier quoted context omitted.

[flagged]

I don't see a link to Elevenlabs. So I'll post one: https://elevenlabs.io/ . It doesn't look open source to me.

Thanks for the link.

Not sure why my question got downvoted. We were talking about Abogen which is FOSS.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#38

Post by logicprog »

I've been using this to try to make audiobooks out of various philosophy books I've been wanting to read, for accessibility reasons, and I ran into a critical problem: if the input text fed to Kokoro is too long, it'll start skipping words at the end or in the middle, or fade out at the end; and abogen chunks the text it feeds to Kokoro by sentence, so sentences of arbitrary length are fed to Kokoro without any guarding. This produces unusable audiobooks for me. I'm working on "vibe coding" my own Kokoro based tkinter personal gui app for the same purpise that uses nltk and some regex magic for better splitting.

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#39

Post by frumiousirc »

This needs to be run from an environment where `pip` is available as that tool is used during the running of the abogen app. Using `uv tool run abogen` gets you started but then the app hangs at model install time. `uv venv && uv pip install pip && source .venv/bin/activate && abogen` lets it run properly.

Otherwise, it's a nicely packaged GUI. Well done!

I tried a PDF and the UI to select pages or sections is good and generation is fast on my laptop's GTX 1650.

The result is an .ogg audio and .ass subtitle file. Played with mpv allows listening and reading along in the terminal. Only issue I have with the result is that visual line breaks from the PDF are preserved resulting in long pauses "randomly" in the middle of sentences. This greatly interrupts understanding of the audio.

Edit: enabling the skipping of single newlines helps!

Re: Abogen – Generate audiobooks from EPUBs, PDFs and text

#40

Post by tummler »

jamilton wrote: post #12
anotherpaul wrote: post #7

Does it turn it into spoken word or an audiobook? Because good audiobooks often have voice actors that read the characters with different emphasis and dialects. I imagine tools like chatgpt could do this for a few sentences but what about an 8-20 hour audiobook? I think there are still basic hurdles to take before we can go epub to audiobook in a quality that can compete with current state of the art. Or am I missing…

Elevenlabs has a feature for a "full cast"-type generation, where different characters will get different voices. It's certainly not automatically sensitive to dialect though. It's probably possible with current systems to do though. I believe there are TTS systems that can use context/prompting to change emphasis and other speech qualities, though I'm not sure how reliably.

I’m sure it’s doable. I think you’d want to break it into a few discrete steps for the best quality. First process the book and identify key info like genre, tone, etc. Use that to determine the best voice(s) and reading style, assign actors for multiple characters/subjects. Maybe output some examples to spot check for approval. Tweak based on that then generate the audio. Prob a couple other steps in there and maybe a bit of custom work to optimize in key areas. If someone wants to do this as a side project I can help scope the architecture and process but I don’t want to code it. :p
Post reply on HN