I just listened to this AI generated audiobook and if it didn't say it was AI, I'd have thought it was human-made. It has different voices, dramatization, sound effects... The last I'd heard about this tech was a post saying Stephen Fry's voice was stolen and replicated by AI. But since then, nothing, even though it's clearly advanced incredibly fast. You'd expect more buzz for something that went from detectable as AI to indistinguishable from humans so quickly. How is it that no one is talking about AI generated audiobooks and their rapid improvement? This seems like a huge deal to me.
A lot of people just aren't aware of how fast AI is moving. AI voices were pretty meh earlier this year. A lot of people working on the audiobook/voice acting scene have been talking about this though.
I recommend everyone to check the YouTube channel "two minute papers" who have being doing videos about papers on AI for the last 10 years on so to see the accelerated progress AI have. Like 5 years ago those images generating AI looked like LSD infused dreams and now they look almost perfect.
Ah yes, Audio AI. I can't wait for this rapidly-approaching future where you literally won't be able to trust the validity of anything your senses tell you anymore
You know some people are just gonna generate that fucking locker room smell, the reek of hormones and axe body spray, to terrorize people
I want TTS made better with AI so that I won't need huge audiobooks filling up my phone. The epubs that I already have would serve as audiobooks when needed.
As someone who only consumes books in audiobook form this is great news for me, I tried to listen to some automatically generated audio books around 2 years ago and I found them horrible to listen to just because they sounded so off.
I'd love to be able to copy in the text of a book and get actually listenable (is that a proper word?) audiobook out of the other side for some books that will just simply never be recorded by actual people due to being too old / obscure.
I've been wanting to be able to listen to the Pelucidar books for years but they just don't exist in audio format, is there somewhere publically available that I can do this?
Look at the description of the video. It's not automatically generated. He made several voices and narrator and applied it to each character.
While insanely cool, it's not "put in book here, get audio book there"
Yes I realise that and was over simplifying in this response but as I stated in another comment I would be more than happy to work on prompts for myself if it could generate something satisfactory to listen to.
The video posted by OP still sounds a bit "dead" so I don't think the tech is quite there yet but it is promising for the future the way it is headed.
That sounds pretty cool, though I'd be concerned it will suffer from the classic problem of current AI (...and humans, but that's by the by) of confident incorrectness. Like an automatic transmission can miss meanings and types of context that a human will spot, programmatically generating speech can probably mess up punctuation and flow - even the way a human reader sometimes will get part way through a sentence and realise they need to start again for it to come out right.
That said, I can't see it being a big problem for most works, just unfortunate here and there. For once it seems an AI application short on downsides! (Except for the usual economic ones for many people previously trained in the field.)
I’ve been getting into audiobooks in a big way recently. This is interesting but somehow seems off to me. Maybe I’ll try listening to one and have my mind changed. We’ll see!
There are also a few AI sung songs out there that are pretty good. Most of them sound pretty Autotuny, but to some extent, that can be a style. Aura, by Ghost, is a good example. If I didn't know it was ai, I would just think it was autotune.
Isn't Ghost and Pals just vocaloid? If I'm thinking of the right ghost(the one who's songs get used in fandom animation videos). That's already been a thing for a while.
That song specifically uses Solaria, which markets itself as "AI" but I'll admit I'm not 100% sure that isn't just marketing https://www.eclipsedsounds.com/product-page/solaria-synthesizer-v-ai
This is amazing. I'm the future, I''d like to try this on old books I've read in the past just to check
It sounds like a generative model to me, but it's probably the best one I've ever heard. Also, thanks for the link! I added it to my listen list!
