- cross-posted to:
- [email protected]
- [email protected]
- [email protected]
- cross-posted to:
- [email protected]
- [email protected]
- [email protected]
You must log in or register to comment.
AI voice synth is pretty solidly-useful in comparison to, say, video generation from scratch. I think that there are good uses for voice synth — e.g. filling in for an aging actor/actress who can’t do a voice any more, video game mods, procedurally-generated speech, etc — but audiobooks don’t really play to those strengths. I’m a little skeptical that in 2025, it’s at the point where it’s a good drop-in replacement for audiobooks. What I’ve heard still doesn’t have emphasis on par with a human.
I don’t know what it costs to have a human read an audiobook, but I can’t imagine that it’s that expensive; I doubt that there’s all that much editing involved.
kagis
https://www.reddit.com/r/litrpg/comments/1426xav/whats_the_average_narrator_cost/
So I produced my own audiobooks for my Nova Roma series so I know the exact numbers for you:
$250 per finished hour for the narrator. Books ranged from about 200k words-270k words, which came out to 22 hours, 20 hours, and 25 hours.
So books 1-3 cost me $5,500, $5,000, and $6,250. I’m contracted for two more books with my narrator, so I expect to spend another 5k-6k for each of those.
So for a five book series, each one 200k+ words, the total cost out of pocket for me will be about $27,000 give or take to make the series into audiobooks.
That’s actually lower than I expected. Like, if a book sells at any kind of volume, it can’t be that hard to make that back.
EDIT: I can believe that it’s possible to build a speech synth system that does do better, mind — I certainly don’t think that there are any fundamental limitations on this. It’d guess that there’s also room for human-assisted stuff, where you have some system that annotates the text with emphasis markers, and the annotated text gets fed into a speech synth engine trained to convert annotated text to voice. There, someone listens to the output and just tweaks the annotated text where the annotation system doesn’t get it quite right. But I don’t think that we’re really there today yet.
The annotated text idea could work but I’m just sceptical of whether or not you would end up doing more work annotating all of the text, listening to it back, redoing certain bits and then editing the final result into a single file then you would if you just had a human do it.
After all you’ve really automated is the reading of the text, which in the grand scheme of things doesn’t take all that long.
No publisher is going to pay a professional to narrate their audiobooks when they can have AI do a shitty job for much less.
A shitty narrator can get me to hate a book I like. A great narrator can bring the characters to life, enhance the experience, and turn me from a listener to a fan. I’ve searched for books by narrators like Nick Podehl and Jeff Hayes and bought audiobooks I wouldn’t have otherwise.
I tried, and failed, to get into audio books for years. Then I listened to Dungeon Crawler Carl narrated by Jeff Hayes and what an absolute delight it was. There’s no way I would’ve gotten even 10 minutes in if it was one of those soulless AI voices instead.
Currently listening to the first book.
I’m not sure why AI would automatically mean it’s doing a shitty job.
Because… the tool has no understanding of anything? It reads written words, yes, but no intention, no cultural context, no intonation. Unless everything is spelled out like a script, then it will not sound great, would it?
Someone can manually go through it and correct and edit it, as one would a regular, human made recording. It’s not rocket science exactly. It wouldn’t be a story time for children but it would probably be alright for more plain stuff
If the “fix” for an AI implementation in a use case is, again, to manually correct it and find a less demanding audience then… yes, by definition it’s shitty.
The point isn’t that it’s infeasible, just that it will be low quality.
I mean you have to correct and edit human made stuff too, doesn’t mean it’s shit lol
If you want the stuff read out and don’t care for the radio type stuff, I’d imagine the better voice AIs do a pretty good job. And I personally prefer the more neutral voices to the story time stuff, so works for me.
This is me just speculating here but if they follow the path of this CEO who fired his human staff to replace it by AI… then rollback admit it’s shit https://gizmodo.com/klarna-hiring-back-human-help-after-going-all-in-on-ai-2000600767 then my bet is that it’s not done to improve quality but rather margins.
If AI is done alongside professionals, and done so ethically (not stolen training data, not ignoring ecological cost by pumping water in dry areas to cool down GPUs, etc) and economically (i.e. not having it “cheap” now but once a monopoly position is obtain, raise prices for a captive set of consumers) then yes it can be potentially empowering. This though is pretty much never the case.
That being said, if one “just” want read aloud, there are plenty of FLOSS alternatives and I believe Mozilla even a TTS/STT system based solely on voluntary voices.
It’s a company, of course it’s done to increase profits. I’m just saying it being AI doesn’t automatically mean it’s shit, it could be done just fine. AI is a tool, the end result depends on how that tool is used.
These people just want to hate AI. Read through and see how many times they complain about copywrited material stolen, but claim piracy is the solution.
Honestly audible are terribles. They are constantly doing things that annoy me, like they must have a team somewhere that spends its days going, how can we kill this golden goose?
They are going through and replacing audiobooks recorded in the 1980s with new ones which in theory should improve their quality but they’re getting rid of the classic sounds of those books.
like they must have a team somewhere that spends its days going, how can we kill this golden goose?
I wouldn’t put it past Bezos to have an actual enshittification department.
Nick Podehl is such an amazing narrator. The voices and performance are amazing.
I’ve been slowly getting through the Kel Kade books and the narration just makes it for me
The thing with this is that there won’t be shitty narrations any more. Hate it all you may, fact of the matter is that AI-powered voice generation is pretty good at what it does. So in the future you won’t have shitty narrations and great narrations. You’ll have decent narrations and great (human) narrations.
And teslas will have full self driving tomorrow and crypto currency will replace normal currency within one year! Always believe in the hype!
That depends entirely on how profitable it is and how much they can get authors onboard.
I do agree that a good narrator delivers a performance that adds the work. James Marster will always be Harry Dresden in my head.
That depends entirely on how profitable it is and how much they can get authors onboard.
A. Anything can be profitable when the cost to generation will be counted in singles of dollars instead of multiple thousands for a good narrator. They don’t even have to sell many to turn a profit too.
B. You think authors are going to have a choice? Lmfao. It’s the publishers that hold any real power and they will jump all over everyone’s IP with AI slop to make an extra three cents.
It’s the publishers that hold any real power
It might be time to finally change that, especially considering what a piss poor job they have been doing for decades at their own part of the production of media.
Your view seems to be hyper focused on the most pessimistic way of interpreting things. Are you doing OK? Seriously, I know how easy it is for everything going on to overwhelm you with negativity. How are you doing?
Maybe this is a culture clash thing, but FWIW, to me your post comes across as incredibly condesending asking a total stranger about their mental helth and implying its bad like you were their close friend.
I find the constant stream of people hyper focused on the worst possible outcome tiresome and frustrating. But instead of responding with that, I intentionally tried to express compassion and concern for a complete stranger. But because this is the Internet, naturally people interpret my actions with the worst possible intent.
That being said, how are you doing? Have anything fun you are looking forward to?
So despite me giving my opinion that that style of posting seems (to me) to be condesending you decided to apply that same style of message, which i just said I thought was invasive, to me?
I get you think you are being nice but trying to force unearned intimacy comes off as creepy.
This isn’t the worst possible outcome. It’s the most realistic one. The worst one would be publishers just straight up replacing their writers with AI
For fiction, yeah, that’s true. For nonfiction, this could work pretty well.
I’m still generally opposed to it because it’s using the work of existing voice recording without compensation, though.
nonfiction, this could work pretty well.
Only in rare cases.
If you have for example some explanations to a complex topic, then a super emotionless voice would still make you hate it and block you from learning it. Even the most dry and hard topics need some good and alive voice in explanations.
If it is just some reference list, where you need to search and hear small parts of it, then it could be Ok.
Maybe we’ll start reading again.
There is literally zero shame in someone consuming audiobooks, and it’s deeply weird to act like something is lost to you if others enjoy them. And this is coming from someone who virtually never listens to audiobooks.
I never said there was. I offered an alternative. . Outrage is misdirected and it’s by design. There are constructive ways to direct it
Reading is not an alternative to listening. Both have different use cases. You cannot read while driving, to name just one.
“Maybe we’ll start reading again” obviously implies that something is lacking presently and that with luck, we’ll go back to the way things were
Not sure if you’re saying I’m outraged but I promise you I’m not, just thought it was lame to try and imply audiobook enjoyers were somehow less than because of how they prefer to enjoy stories
I made some AI animated content that I never released because I don’t have the rights to the voices I was using. Even though I was blending several voices together to make them unrecognizable, it made me uncomfortable.
But in the process I learned the capabilities and limitations of AI voices. If you’re going purely from text to speech, it’s horrendous (as far as I experienced). Very robotic. It’s a bit better when melodic information is included (as in Suno) but still sounds like AI.
But when I recorded my own voice saying the lines and then converted it to another voice, it took all of the nuance of my line reads and converted it into the other voice.
So, would your opinion change if it turns out they’re going to use purchased voice rights to have a single narrator perform the whole book and then use AI to turn the narrators voice into a full voice cast?
I could see how it would allow lesser known books to have a better experience with a truly separate voice for each character, but I could also see how this might drive out lesser known/minority voice actors. Not advocating one way or another, just providing a piece of this conversation I think we should bear in mind.
So, would your opinion change if it turns out they’re going to use purchased voice rights to have a single narrator perform the whole book and then use AI to turn the narrators voice into a full voice cast?
It would make me hate it even more because I already hate the existing full cast of humans audio dramas 99% of the time and actually prefer a single (or low number of) narrator approach.
Completely fair. I kind of like them. They did it for Redwall and I listen to those books on long drives sometimes. It works for me. Now I guess the advantage could be to have both versions and get to choose which you listen to–but even I’m skeptical that a corporation would have that much regard for the preferences of its consumers.
Oh. That’s an interesting use-case I hadn’t considered.
Using different voices to read different parts of a book turns an audiobook into a bad audio play, and arguably, a bad audio play is worse than a mediocre audio book.
What audible misses is, that, while reading is a technique that can be automated, narrating is an art. They can use AI to read books, they cannot use AI to narrate books.
Your example of AI use is a good example of this: AI can read your content. AI can enhance your capabilities. But only you can narrate it.
A shitty narrator can get me to hate a book I like.
And that is where I see potential for AI. There are quite a few books which I’d love to listen to but they are all narrated by a guy whose narration I can’t stand. AI would open the possibility to choose a voice and I might actually get to enjoy those books. It’s Amazon though so the ethical implications and quality concerns are something I’m worried about.
Did you ever heard a single AI-narrated content that did not make you run away screaming?
You think they’ll be narrating books with Tiktok TTS?
To rephrase my question: where can I listen to an example of good AI spoken content?
First thing that comes to my mind would be Dougdoug. He’s a streamer who messes around with AI a fair bit for funny content, including using AI-generated voices at times.
Hm. I wasn’t able to listen to all 9:53:57, but in the samples I watched I heard a voice resembling the classical computer voice of Science Fiction movies of the 70s. Better than most YouTube AI generated audio content, but good enough to narrate audio books? Well, we’ll accustom to anything, I guess.
I don’t know if my timestamp went through, but the part I linked to was at 7h/19m/42s. That’s the relevant part, not necessarily the entire video. That’s a showcase of good AI voices.
Some use even worse, if YouTube content is any indication.
But you think Audible would use those to narrate books?
I think an Amazon owned company would actually default to Alexa, who’s voice is equally terrible. Why pay more to develop a better voice?
deleted by creator
Beautiful, it works. Why not.
Left Amazon a handful of years ago. Glad I didn’t entirely contribute to this. Saw that coming….
Surely I can just do that myself with an an epub and a free AI.
Glad I binned my Audible subscription many years ago.
It’s Amazon, what did you expect? Enshittification and monopoly abuse, no surprise.
Idk, they have pretty good stats that nobody will listen to an audio book if they don’t like the narrator, so being able to choose your own narrator on the fly isn’t really shitty
Enshittification isn’t adding new features that people want, it’s gradually lowering the quality of the product. So here if Audible is solely adding more possibilities, never at the cost of higher quality ones degrading, then indeed I’m wrong.
If though they hire less people to do good voice acting, then it’s really shitty.
I genuinely hope I’m wrong and they are ONLY adding new capabilities… but my entire experience with capitalism is that obtaining a monopolistic position is not done to improve quality but rather to increase margins regardless of how.
We’ll see!
More jobless, desperate people.
AI will write them and AI will read them to us.
that’s gross.
Let AI pay for them and AI listen to them too. That way we can pay for and listen to actually good ones.
Stock up on old physical books
It is easier to keep the books than what’s written in them…
For now at least I bet this’ll be pretty mediocre. I’m a big audiobook fan and voice actors have a massive impact on the quality of the finished product. A great voice actor can make a mediocre book fun and engaging, a bad one can make a great book unlistenable. The best do great voice differentiation. As an example I’ve really enjoyed Andrea Parsneau’s work in The Wandering Inn series.
Imagine not liking the voicing of a book, so you just pick a different one.
You seem to be implying that’s ridiculous, but it is indeed exactly like that, though it’s not like I’m expecting every performance to be a masterpiece.
It’s also pretty subjective, for example folks either seem to love or hate R. C. Bray. My mother can’t stand the guy’s style, I think he’s okay.
No, I think it’s great to be able to get rid of shitty voice work with the click of a button. Wish I could use it on my bf’s Brian Sanderson audiobooks. That guy’s simpering, exaggeratedly high pitched female voices are so unpleasant to listen to.
Ah, I see what you’re saying, I misunderstood and thought you were taking about picking a different book. Indeed, for the worst case scenario a mediocre AI voice could be an improvement!
Patrick Tull’s Aubrey/Maturin series is fucking amazing.
I listened to one recently that was using AI. It was kind of off putting because of how robotic it came off.
It wasn’t the tone really, but I find that AI tends to not get human speech inflections right most of the time during active speech. And that can be jarring to me at least.
I am okay with this only in cases where 1) the author approves, and 2) there is no audible version anyways.
Some people prefer listening to their books instead of reading and that’s totally ok. Indie authors can’t always afford to hire a narrator but I’d still want the buyers to be able to listen to the book.
Big question is, will the author get paid for the download or not…
I wouldn’t support it even if the author couldn’t afford it otherwise. There’s no test to confirm that and knowing profit margines, all publishers will use AI for all their books.
Yes, I’d want smaller authors to have people listen to their books, but without oversight, it’s going to ruin all audiobooks.
Save a profile in tts server, then go into read > tts settings and change voice to profile you saved. I don’t remember but you may need readera premium.
youtube already does it.
And it’s shit
YouTube is crawling with it. It’s unlistenable shit. The prosody is badly implemented, pronunciation is infuriatingly bad, and a lot of the text that these TTS are reading appears to be AI-generated. Otherwise, already dire standards of literacy are getting worse at an accelerating rate.
like how fans got obssessed with AI generated DBZ(what ifs)
tiktok voice:
hate. let me tell you how much i’ve come to hate you since i began to live. there are 387.44 million miles of printed circuits in wafer thin layers that fill my complex…
unironically, that is a character that could use an uncanny robotic AI voice.
The professional ai voices are amazing
I will Avoid Audio books with AI voices. I prefer the warmth of a human voice instead. A lot of my favorite Audiobooks are elevated by the interpretation of the professional actor. I am glad I have never even touched the Audible service, now I never will. I will also never consume anything written by a fucking AI.
I wish I didn’t have to go with audible, I am acutely aware that I don’t really own the audiobooks and they could be taken off the platform at any time.
I wish there was some other option but I don’t know of any other audiobook providers. They have the industry locked down.
