3 ms·

Superhuman TTS is well within the capabilities of the big AI labs. Even Google had voice indistinguishable from human back in 2017, but they deliberately kneeca

Superhuman TTS is well within the capabilities of the big AI labs. Even Google had voice indistinguishable from human back in 2017, but they deliberately kneecapped it because of the potential for misuse. Boomers and older folks are not culturally or mentally equipped to handle it - even the crappy open source voice cloning we had in 2019 got used to scam people into buying gift cards. Because of the potential for abuse, nobody wants to release a truly good, general model, because it makes lawyers lose sleep. A few more generations of hardware, though, and there will be enough open data and DIY scaffolding out there to produce a superhuman model, and someone will release it. Deepfake video is already indistinguishable from real video (not oneshot prompt video generation, but deliberate skilled craft using AI tools.) Higgsfield and other tools allow for spectacular voice results, but it takes craft and care. The oneshot stuff is deliberately underpowered. OpenAI doesn't want to be responsible for a viral pitch-perfect campaign ad, or fake scandal video sinking a politician, for example. Once the lawyers calm down, or we get a decent digital bill of rights that establishes clear accountability on the user of the tool, and not the toolmaker, things should get better. Until then, look for the rogue YOLO boutique services or the ambitious open source crew to be the first to superhuman, widely available TTS.

fortran77 1y ago
Stop blaming us old people for your lack of good TTS models.
- onedognight 1y ago
  Username checks out.
bigfishrunning 1y ago
> Boomers and older folks are not culturally or mentally equipped to handle it I think a lot of younger people are also not mentally equipped to handle it. Outside of the hackernews sphere of influence, people are really bad at spotting AI slop (and also really bad at caring about it)
- tonyarkles 1y ago
  Half tongue in cheek when I say this… that might be true, but what are the odds of them actually answering a phone call?
- bsder 1y ago
  > people are really bad at spotting AI slop Erm, guilty as charged? Although, I don't think you can blame people for that. There was a video recently comparing a bunch of "influencer girls" that had signs of "This is AI" and "This is Real". They could all have been AI or could all have been real. I have zero confidence that I could actually spot the difference. This is doubly true as an "Online Video Persona" has a bunch of idiosyncrasies that make them slightly ... off ... even if they're real (example: YouTube Thumbnail Face, face filters, etc.). AI is really good at twigging into those idiosyncrasies and it serves as nice camouflage for AI weirdness.
- tatersolid 1y ago
  This isn’t true in my experience. I’m a parent of three teens and they immediately exclaim “clanker” and hit skip when encountering any form of AI generated content on any platform. They recognize AI slop easily and definitely do care enough to avoid it. As do their friends. AI-generated content has near-zero commercial value long-term.
  - 8n4vidtmkvmk 1y ago
    I'm very skeptical about this zero commercial value claim. I don't think everyone skips past it, and even if they do, that's just the stuff they're detecting as AI. How much have they not identified? What about in a couple years? Heck, even humans subtly trying to sell something give off a vibe you can pick up quickly. But now and then they're entertaining or subliminal enough that they get through.
munk-a 1y ago
> Boomers and older folks are not culturally or mentally equipped to handle it I'm glad you mentioned this because the "Grandma - I was arrested and you need to send bail" scams are already ridiculously effective to run. Better TTS will make voice communication without some additional verification completely untrustworthy. But, also, I don't want better TTS. I can understand the words current robotic TTS is saying so it's doing the job it needs to do. Right now there are useful ways to use TTS that provide real value to society - better TTS would just enable better cloaking of TTS and allow actors to more effectively waste human time. I would be perfectly happy if TTS remained at the level it is today.
- 8n4vidtmkvmk 1y ago
  I still think it would be fun for a video game. Write a backstory for a whole bunch of NPCs and let the player dig as deep as they like. I'm not sure what the bottle neck right now is. Either this idea isn't as fun as I think or we can't do it in real time on consumer hardware yet.
  - munk-a 1y ago
    A lot of full conversion mods just find community members that want to do some VO for practice or as a resume booster or just for the funsies. I think you'd be surprised how easy it is to get half-decent voice actors if you've got an interesting idea to build out.
    - hdjrudni 1y ago
      The problem isn't hiring/paying the voice actors, it's that the NPC can say anything. It's not pre-scripted.
jsheard 1y ago
What does it even mean for a TTS model to be "superhuman" when the goal is to imitate human speech?
- observationist 1y ago
  A single model that can produce voices indistinguishable from human speech, cloning any voice perfectly, and produce audio faster than a human can. Superhuman specifically in the capability to be faster - minutes of pitch perfect voice per second of operation, for example.
rurban 1y ago
The movie industry would love to use it for dubbing. And pay a lot. They think they pay too much for the speakers already.

fortran77 1y ago

Stop blaming us old people for your lack of good TTS models.

onedognight 1y ago
Username checks out.

bigfishrunning 1y ago

> Boomers and older folks are not culturally or mentally equipped to handle it I think a lot of younger people are also not mentally equipped to handle it. Outside of the hackernews sphere of influence, people are really bad at spotting AI slop (and also really bad at caring about it)

tonyarkles 1y ago
Half tongue in cheek when I say this… that might be true, but what are the odds of them actually answering a phone call?
bsder 1y ago
> people are really bad at spotting AI slop Erm, guilty as charged? Although, I don't think you can blame people for that. There was a video recently comparing a bunch of "influencer girls" that had signs of "This is AI" and "This is Real". They could all have been AI or could all have been real. I have zero confidence that I could actually spot the difference. This is doubly true as an "Online Video Persona" has a bunch of idiosyncrasies that make them slightly ... off ... even if they're real (example: YouTube Thumbnail Face, face filters, etc.). AI is really good at twigging into those idiosyncrasies and it serves as nice camouflage for AI weirdness.
tatersolid 1y ago
This isn’t true in my experience. I’m a parent of three teens and they immediately exclaim “clanker” and hit skip when encountering any form of AI generated content on any platform. They recognize AI slop easily and definitely do care enough to avoid it. As do their friends. AI-generated content has near-zero commercial value long-term.
- 8n4vidtmkvmk 1y ago
  I'm very skeptical about this zero commercial value claim. I don't think everyone skips past it, and even if they do, that's just the stuff they're detecting as AI. How much have they not identified? What about in a couple years? Heck, even humans subtly trying to sell something give off a vibe you can pick up quickly. But now and then they're entertaining or subliminal enough that they get through.

munk-a 1y ago

> Boomers and older folks are not culturally or mentally equipped to handle it I'm glad you mentioned this because the "Grandma - I was arrested and you need to send bail" scams are already ridiculously effective to run. Better TTS will make voice communication without some additional verification completely untrustworthy. But, also, I don't want better TTS. I can understand the words current robotic TTS is saying so it's doing the job it needs to do. Right now there are useful ways to use TTS that provide real value to society - better TTS would just enable better cloaking of TTS and allow actors to more effectively waste human time. I would be perfectly happy if TTS remained at the level it is today.

8n4vidtmkvmk 1y ago
I still think it would be fun for a video game. Write a backstory for a whole bunch of NPCs and let the player dig as deep as they like. I'm not sure what the bottle neck right now is. Either this idea isn't as fun as I think or we can't do it in real time on consumer hardware yet.
- munk-a 1y ago
  A lot of full conversion mods just find community members that want to do some VO for practice or as a resume booster or just for the funsies. I think you'd be surprised how easy it is to get half-decent voice actors if you've got an interesting idea to build out.
  - hdjrudni 1y ago
    The problem isn't hiring/paying the voice actors, it's that the NPC can say anything. It's not pre-scripted.

jsheard 1y ago

What does it even mean for a TTS model to be "superhuman" when the goal is to imitate human speech?

observationist 1y ago
A single model that can produce voices indistinguishable from human speech, cloning any voice perfectly, and produce audio faster than a human can. Superhuman specifically in the capability to be faster - minutes of pitch perfect voice per second of operation, for example.

rurban 1y ago

The movie industry would love to use it for dubbing. And pay a lot. They think they pay too much for the speakers already.