Launch HN: Play.ht (YC W23) – Generate and clone voices from 20 seconds of audio

459 points by hammadh ↗ HN
Hey HN, we are Mahmoud and Hammad, co-founders of Play.ht, a text-to-speech synthesis platform. We're building Large Language Speech Models across all languages with a focus on voice expressiveness and control.

Today, we are excited to share beta access to our latest model, Parrot, that is capable of cloning any voice with a few seconds of audio and generating expressive speech from text.

You can try it out here: https://playground.play.ht. And there are demo videos at https://www.youtube.com/watch?v=aL_hmxTLHiM and https://www.youtube.com/watch?v=fdEEoODd6Kk.

The model also captures accents well and is able to speak in all English accents. Even more interesting, it can make non-English speakers speak English while preserving their original accent. Just upload a non-English speaker clip and try it yourself.

Existing text to speech models either lack expressiveness, control or directability of the voice. For example, making a voice speak in a specific way, or emphasizing on a certain word or parts of the speech. Our goal is to solve these across all languages. Since the voices are built on LLMs they are able to express emotions based on the context of the text.

Our previous speech model, Peregrine, which we released last September, is able to laugh, scream and express other emotions: https://play.ht/blog/introducing-truly-realistic-text-to-spe.... We posted it to HN here: https://news.ycombinator.com/item?id=32945504.

With Parrot, we've taken a slightly different approach and trained it on a much larger data set. Both Parrot and Peregrine only speak English at the moment but we are working on other languages and are seeing impressive early results that we plan to share soon.

Content creators of all kinds (gaming, media production, elearning) spend a lot of time and effort recording and editing high-quality audio. We solve that and make it as simple as writing and editing text. Our users range from individual creators looking to voice their videos, podcasts, etc to teams at various companies creating dynamic audio content.

We initially built this product for ourselves to listen to books and articles online and then found the quality of TTS is very low, so we started working on this product until, eventually we trained our own models and built a business around it. There are many robotic TTS services out there, but ours allows people to generate truly human-level expressive speech and allows anyone to clone voices instantly with strong resemblance. We initially used existing TTS models and APIs but when we started talking to our customers in gaming, media production, and others, people didn't like the monotone robotic TTS style. So we doubled down in training a new model based on the new emerging architectures using transformers and self supervised learning.

On our platform, we offer two types of voice cloning: high-fidelity and zero-shot. High-fidelity voice cloning requires around 20 minutes of audio data and creates an expressive voice that is more robust and captures the accent of the target voice with all its nuances. Zero-shot clones the voice with only a few seconds of audio and captures most of the accent and tone, but isn’t as nuanced because it has less data to work with. We also offer a diverse library of over a hundred voices for various use cases.

We offer two ways to use these models on the platform: (1) our text to voice editor, that allows users to create and manage their audio files in projects, etc.; and (2) our API -

476 comments

[ 120 ms ] story [ 8135 ms ] thread
This is the only platform that seems to be offering unlimited voice generation for a fixed monthly price. Does this have a real-time streaming option?
play.ht user here! Awesome service and thanks to you guys I made my first 50$ while generating voices and using them in a short explainer video.
nice voting ring you got there
Are the founder's names causing you panic? can't you appreciate a good startup?
Huh? What do the founder's names have to do with anything?

Personally, this sounds like an extremely irresponsible startup, but I don't know much about it so I'm trying to reserve judgement.

I had to re-read your comment four times before I even understood what you were on about.

I'm mildly agitated by your comment, so I'd like to take the liberty of pointing out that you are the only one in this entire thread linking the names Hammad and Mahmoud to racism. Everybody else in the entire thread is talking about the product on its merits. There's a heated debate and nobody gives a fuck about where the founders are from. That's how it should be. Stop making the world a worse place than you found it.

And, FWIW, I think that the product looks pretty neat. And that the voting ring was just too obvious a play :-)

How do you assert that the cloned voice has been truly permitted by the voice owner? I've had my voice cloned without my consent by other people using Descript and Eleven Labs.

What is your process for verifying consent?

It's mentioned in the second demo video that they have a strict process to prevent cases like yours. I think Descript started asking for identity verification after its service was abused. This one probably has a similar process too.
(comment deleted)
I think the previous comment wants to know what the "strict process" is exactly.
Right, and I'm sure their "strict process" is something like "we take it down after you notify us and provide proof that the voice is yours".
When I tried this service previously, you had to read (out loud) something saying that you were giving consent.
I'd be curious what the false positive rate on that is. Can you clone anyone's by collecting a set of ten voices with unique timbre reading the required statement plus pitch control to get close enough? A hundred? Or can you trick the neural net by giving it something that sounds like white noise to humans until the NN triggers in the right way and goes "ok yep that's a match, you're authorised now"?

Probably not something we'll get to hear as part of the PR pitch.

Or is the consent statement the thing that will be cloned and is there no separate training audio? Then it might actually work and you'll just have to get close enough that the human you're trying to fool can't distinguish anymore (defeating the need for this tech in the first place, at least in targeted rather than automated cases).

Yeah, good point - don't know. When I tried I actually did get a (personal?) email saying that it didn't match closely enough. After uploading another sample (based on a different text) it went through.

I like your idea of just training on the consent text! That wasn't the case when I tried it as you needed around 3h (optimally) of training data.

If someone has the capability to trick the service like that, they likely have the capability to recreate the functionality themselves.
With a couple soundalike voices and changing the pitch in Audacity? That's a far, far cry from doing cutting edge neural networks that clone voices with samples of less than half a minute.

If you mean the white noise, I meant that as a brute force attack because, to do it more targeted (to know what it'll accept as seeming like your target voice), you'd likely need their exact model rather than doing your own.

(comment deleted)
TIL, the Booth Junkie is on HN. Love your work, sir.
Remarkable, I spoke spanish on the training audio without realising. Then every two options were one with a latino-accented english and one with an indian-accented english
Well, nothing could possibly go wrong, eh?

If your homepage is toadying up to Musk by claiming he has "limitless intellect" then I've already heard enough.

We have a duty to consider how what we build can be used to harm others. If the obvious and many ways this could be abused aren't covered anywhere I can find on your website, then I'm going to conclude you haven't considered them. Which is terrible.

yeah, we should just make all AI research illegal. I mean, without gate keeper the world will fall apart! Did you know that you can draw and write _anything_ with a pencil! We need to get onto that next. And the internet, you can publish anything you want on there!
You seem to be intentionally missing the point.
(comment deleted)
Don't be so sure it's intentional
I can’t wait for spoofed messages from my loved ones.
The scam via voicemail possibilities are endless!
Just in time for April fools day.
This is actually something I would use myself. I've checked out the few AI voices but they are not of the same quality. Majority of them sound still very robotic and not like the real person. The only passable one for me was "William"
Congrats on the launch, your text to speech quality is unparalleled.
https://play.ht/app/voice-cloning > Clone a voice now

Pops a modal: Try Voice Cloning for Free!

Enter a credit card for $0.00/mo with no other information on screen

Bounce.

Why not let me play around with it a little without asking for a credit card?

I think if you are cloning voices, you should be required to have a credit card or some other KYC identifier. Even if it's free. This kind of highly abusable tech should have a paper trail IMO.
As someone whose voice has been cloned without my consent, I could not agree more.
I guess I misunderstood/didn't think it all the way through. Not sure what the balance should be but... I just wanted to see how it would be at cloning my voice (not "a" voice that doesn't belong to me) as a quick gauge to "is this technology ready to play around with".
yeah because that's working great for crypto lol
What do you mean? KYC is required on every US exchange.
Exactly, that's why the crypto space doesn't have any scams
It's trivial to get your money to exchanges run by people/machines who don't care to comply with US law, or to render the KYC worthless in the first instance.
It's an effort to prevent abuse. We previously asked users to pay upfront but most people want to try it out first.
Do you accept anonymous Visa/Mastercard/etc. gift cards in this payment method? If you do... are you actually preventing abuse or just making it slightly more complicated to pull off?
I would mention something to that effect in the modal because it wasn't clear to me why it was asking for card details at that point for "$0.00/mo" (though I guessed the reason). Maybe something like "To prevent abuse, we require card details, but you won't be charged", but worded better.
> but you won't be charged

"no matter what based on your usage/you are locked into the free tier" would have helped for sure

i still would've bounced because i just wanted to goof off with it quickly while it captured my attention and requiring a payment method is just... terrible friction to capture users being able to quickly test one of the key features you advertise, but i guess if fraud concerns are that bad, that's the tradeoff you have to accept?

Thank you. Would fix this.
This is the first startup here where I think the tech should essentially be illegal.

It's cool tech, yes I'm impressed at the achievement. Nuclear weapons are impressive too.

OTOH this kind of thing is getting easier and easier to do, so what's a realistic way forward?

Our only hope is that politicians and celebrities get sick of their voices and likeness being used to scam people or sell crypto or viagra and get laws passed against this type of impersonation.
actually, this is the way. force their hand.
I guess eventually people will go back to only meeting face to face for important communications. I don't know what the way forward is for news.

I truly do not understand people like these founders, obviously they understand the future they're creating. "If not us, someone else would do it" is not an excuse. Neither is "I like money".

As this is what every "organized crime" (people who dont want prying eyes) groups have done for centuries.

Next, normal people will adopt the Mafia's 'cover the mouth while pretending to use a tooth-pick whilst talking to prevent lip reading from remote viewers (same thing sports people do currently.

-

My granmother was deaf for the latter half of her life. She became an expert lip reader.

It was fun going to restaurants with her as she would tell me what people at the tables far away were talking about "oh that couple isnt having a happy time..."

There are some cool uses like dubbing movies in foreign languages while keeping the original "voice styles" or having your long dead relatives talking to you in some memorabilia etc. It could also cause unexpected creativity explosion e.g. in games or fan fiction movies. To avoid misuses we might perhaps find the only good use of blockchain.
>long dead relatives talking to you in some memorabilia etc.

It seems a bit weird to me though. I mean, looking back at old records can still pass as mere nostalgic behavior. Wanting new sentences pronounced in disguise of lost relative voices is not great in term of respect for these people to share my own feelings.

Also I guess that now there is not much preventing completely new songs with whatever lyrics staring voices of Elvis, Hendrix and Pavarotti. Actually a continuous flow of on the fly generated lyrics seems perfectly plausible at this level, isn't it?

My grandpa wrote a fiction book so having him read it to me would be kinda cool, even if he's long gone. Still, he technically exists in the 4D universe but the time dimension no longer overlaps with mine.
Foreign language dubbing is a great use case. And the ability to alter the video such the the lips are synced to the dubbed version would be a great addition. I can’t believe studios are using these things already (the video part in particular).
I think we're not that far from the day all movies will be produced by AI, including all parts in various languages using the most popular actors per given market, all accurately translated and of course perfectly synced since there would be no dubbing but creation on the fly. First they'll use virtual copies of real actors by purchasing rights from their estate, until the public will slowly accept full virtual and cheaper ones. I give them 20 years max, and I'm being optimistic (pessimistic?).
About the blockchain comment… For years, I’ve been expecting camera makers (including phone makers) to offer image hash verification on blockchain at the moment of image capture. I’m surprised it’s not routine.
1. Expensive

2. Requires internet

3. What image do you verify? Between auto-retouching, manual retouching, compression, filetype conversion, an image file might be invisibly transformed 10 times in between capture and Instagram upload.

4. Useless for disproving fake images until every camera manufacturer in the world has implemented this.

5. Hostile to customers, now your picture doesn't get the green verified badge or whatever if you decide to crop it or something.

Ad 3) all of them. They will be recorded in a cryptographically recorded chain and you'll be able to backtrack all steps.
I'm now wondering how many images are generated a second that would all need to be recorded. How much is it going to cost to take a photo?
I don't think that works because there isn't One True Blockchain and it doesn't seem like there ever will be. How do you get Google, Huawei, some random photo editing app, Dropbox, Meta, etc. to all agree on the same chain?

This is a general problem that I have with a lot of blockchain ideas. For example, there are a few startups that claim to verify carbon offsets by registering them on the blockchain. There are many problems with this that we don't need to get into, but the relevant one here is: what is stopping me from registering the same offset on three different chains?

I guess a mix between impractical and completely useless?
The only thing that blockchain can do that couldn't be done before is cryptocurrencies (not sharing my opinion about them here).

Pretty sure this is not a good use of blockchain, and I don't see how it would remotely avoid misuses.

> I guess eventually people will go back to only meeting face to face for important communications

This seems to be the only realistic future. This sort of technology literally makes it impossible to trust anything electronic.

People were worried about the balkanization of the internet, but now they look like optimists.

PGP exists, it is just no one uses it
PGP doesn't help you with phone and video calls.
not really the point. a PGP-style tech could easily exist for phone or video tomorrow, if it doesn't already. but PGP-style tech for email (called "PGP") has existed for 32 years and basically no one uses it. whether or not the tech exists doesn't matter nearly as much as whether or not people actually use it.
I'm unclear as to what your point is, then...
you said: "This sort of technology literally makes it impossible to trust anything electronic."

we said: "No, because there is also technology that makes it possible to trust anything electronic with very nearly 100% reliability. But no one uses it"

I think your first statement is both technically wrong and generally wrong. Electronic trust is a solved problem...it's just that right now, it's really not as big a deal as some people are worried about it being, so we haven't generally implemented the solution. We could make electronic cars for long time before other things made them commercially viable.

You can short circuit all that shit by merely compromising the device.
trusted execution is also a thing, just largely unused/underutilized. in my opinion hardware/software platforms can be designed such that the only real exploit would be for someone to insert an attack vector into the hardware (IC) itself, which is nation-state level work. again, possible but not used in practice because of the perceived risk-reward tradeoff at the moment.
Yea until the next Vault7 leaks and the "state level work" is accessible to all.
Ahh, I see.

I disagree that electronic trust is a solved problem. It is mathematically solved, yes, but the reason that it isn't widely used is because it's still intrusive and painful to do. A solution that isn't acceptable to the masses isn't an effective solution.

If it could be done in a way that is invisible (like HTTPS, for instance), then it would be ubiquitous. That's the part of the problem space that still needs resolution.

It's called signing.

I send you a bunch of bytes signed with my private key (which somehow you have to verify in a trusted way) and you can be sure that I am the person who signed those bytes (unless I was compromised).

Thanks, off to work on a new startup!
It doesn't have to be PGP specifically to satisfy the goal of strong keypair-based security finally having traction — does it?

USB etc hardware keys can be used protect accounts like Gmail, Coinbase, web hosting services, many more now.

Fun fact: it's possible to receive Facebook user notification emails encrypted against your public PGP key.

There's a book I've been waiting years for the audiobook to come out. Plenty of legitimate uses for this tech. Plenty of horrible ones too. It's the same with many technologies, no? I don't think outright banning it makes any sense.
People still accept contracts based (in part) on scribbles on paper. Fraud will happen, just like it does for signatures. I'm sure sometimes countermeasures will be done (including meeting in person), but it's not like video chat or phone calls will completely disappear.
> face to face for important communications

Funny how computing perfected communication and ultimately will undermine itself.

> I don't know what the way forward is for news.

I'd say every packet of voice/img will have to be signed by the recording device and checked at rendering time.

> I truly do not understand people like these founders

Me neither. Don't do it. We don't need that, and the malevolent use of this will confuse people to an extreme point.

Even the 'good' use of having a deceased relative utter new sentences is beyond strange. This is too far gone. And I'm no luddite.

This is democratising the tech. Otherwise only the intelligence agencies will have it and we will continue to be duped not knowing what is possible.
We don't need to all be able to use the tech for it to be known publicly.

Apply your same logic to any other easily misused tech:

"We must all have easy access to bio-engineered viruses. Otherwise only..."

"We all need to have access to nuclear weapons. Otherwise only..."

Not all tech should be in everyone's hands.

Its a different kind of tech. A society changing tech that can be used surreptitiously. It needs to be in people faces in (for example) the form of over the top and ridiculous memes.

That is not possible with or comparable to, things such as bioweapons.

That is a fair point. It is different in that it could be used without necessarily creating destructive ends. Ultimately though, once the majority of people are aware of the capabilities of the technology, is any good being done by still allowing it to be easily accessible? It seems that the value in spreading its use reduces proportionally to the population of people who are still ignorant of it, while the danger of its misuse rises with number of actors/users until fully effective, equally accessible countermeasures are in place. If that's accurate, then it seems that the plan you're advocating for increases danger as quickly as possible and keeps it high while the world works on mitigations.
The plan I'm advocating for immunise the public against the danger of misinformation through deepfakes, as well as cause a lot of resource to be thrown at the problem as the public would surely be very uncomfortable in such a situation
I stopped trusting the news years ago, between whoring out for engaging but divisive content and obvious political bias, it's been a crapshoot since GWBs cronies gutted the FCC.
But what is a realistic way forward? Do you think that scammers won't have this technology in 2 years? Can we really prevent any illegal use of neural networks at this point? With weapons that you actually have to physically buy, you can intervene on a country level (to some degree). But already with those 3D printed ones, we are basically doomed. Of course it's a tragedy of the commons type of situation. But banning all legal uses does not prevent the illegal ones.
Most scammers are incredibly lazy and honestly not all that competent. There's no need for them to change that if you can prey on the weak and vulnerable.

The difference between "the paper is out there" and "there's a button to do this" is quite obvious in cases like software exploits. A report of finding a vulnerability rarely leads to a massive automated exploitation campaign, but if that report also contains a proof of concept the amount of automated attacks radically increase. I believe the same is true for many other types of crime: even a mild bar to entry will prevent a significant amount of criminals from advancing their techniques.

I think the negative impact of these voice changers is much bigger than the advantage we gain as a society. Criminals will always exist, even crafty ones, but "we can't prevent crime so let's not bother trying to do anything about it" is not a great take in my opinion.

Of course it's a tragedy of the commons type of situation

TOTC is about resource depletion. GYI. It's not applicable here.

I'm assuming you're not talking about Global Youth Impact, so what does GYI mean here? Online dictionaries are not helping.
(comment deleted)
What if the voice sample is somebody saying they give specific consent to be cloned by that service?

You could of course clone a voice to generate that "consent" -- but at that point there's no additional harm done because they'd already have the clone.

It's unrealistic that this tech won't exist somewhere, even if the big actors stay away for ethical reasons. A voice auth practice strikes me as a good compromise.

> You could of course clone a voice to generate that "consent" -- but at that point there's no additional harm done because they'd already have the clone.

But this possibility renders the idea unviable in the first place, does it not?

(comment deleted)
>Introducing the National Postal Service - send a letter to anyone for a nominal fee. No need for a personal courier, armed escort, or patrician status.

>This is the kind of thing that should be illegal. Now, any Plebian could essentially write a letter to anyone, impersonating anyone. Forged letters could drag us into a war with Persia - for Jupiter's sake!

This is a hilariously bad attempt at discrediting the original argument. There's a vast difference between forging a letter and replicating the unique vocal fingerprint of any human being, on demand.

I suppose if we approach the point that we can create robotic clones of anyone, anywhere, that look, sound, and move like anyone on the planet, that will be just like the post office too, right?

What are some of the differences? Besides the glaringly obvious text vs audio. I mean prior to telegraphs if I got a letter from my sweet heart with a lock of hair or something and a request for funds I’d probably believe it, especially if it took days or weeks to communicate back and forth?
Impersonating a letter is similar to having an impressionist record an impersonation of someone's voice. It's difficult, very imperfect, and not very scalable.

The analogy for this technology would be a robot that can perfectly imitate someone's handwriting and vocabulary using one letter as a reference.

No, it's more like those "stress tester" services that you're definitely-certainly-fingers-crossed supposed to only point at your own servers.

Sure, this is marketed as generating your own voice to read scripts for you YouTube channel, but are they actually verifying who's voice you're generating?

You completely miss the point: scale.

Try to deceive people by learning about their contacts, writing a convincing letter and sending it. How long does it take you to prepare one letter?

Now those AIs potentially allow you to just generate millions of those with one click. The problem is the scale: everyone can do it at no cost and at scale.

Making then the illegal would accomplish nothing since it's already out in the wild. You can generate audio with high quality on fine-tuned versions of Tortoise TTS, which was originally trained on a cluster of NVIDIA 3090's, so it's within reach for any smart person to train a from-scratch model on consumer hardware. Realistically? We have to accept that this tech exists and there will be both positive and negative outcomes from it.
> Making then the illegal would accomplish nothing since it's already out in the wild.

Not true. Making it illegal wouldn't make it nonexistent -- that's true. But making it illegal would provide at least some method of mitigating some of the harm.

That's more than what we have right now.

> We have to accept that this tech exists and there will be both positive and negative outcomes from it.

Of course. But that doesn't mean it's futile to try to reduce the negative outcomes.

> Not true. Making it illegal wouldn't make it nonexistent -- that's true. But making it illegal would provide at least some method of mitigating some of the harm.

It's already illegal to impersonate someone to steal money or scam them, and those laws were on the books before computers existed.

> Of course. But that doesn't mean it's futile to try to reduce the negative outcomes.

You can run something on a consumer GPU and it's every bit as good if you know how to dial it in. By the end of the year you'll be able to download a nicely packaged "voice cloner" from a torrent that runs on a cheap laptop. IMHO any effort on regulation is far better spent informing people rather than trying to put the cat back in the bag.

> It's already 100% futile.

I don't think so at all. There are all sorts of things you can technically do with ease that are illegal for good reason. Laws against them aren't futile.

But I admit that perhaps I'm being overly optimistic here. I'm just trying very hard to see any way that this stuff can end up not being a complete societal disaster.

> It's already illegal to impersonate someone to steal money or scam them, and those laws were on the books before computers existed.

There are two hurdles a criminal has to get past:

1. decide to break the law 2. figure out how to pull off their scam

It sounds like you're saying that since hurdle #1 already exists, hurdle #2 is irrelevant? No, of course it isn't. That's like saying that gun control can't possible help because it's already illegal to shoot someone.

Adding difficulty to a crime reduces (but does not eliminate) the prevalence of the crime.

Scale matters. The difference between 50K smart and dedicated criminals being able to use this technique vs anyone with a web browser is significant.
You may get your wish. The FTC posted an article about this a week ago. [1]

> The FTC Act’s prohibition on deceptive or unfair conduct can apply if you make, sell, or use a tool that is effectively designed to deceive – even if that’s not its intended or sole purpose.

It seems like an awfully broad rule? But they probably could go after this startup if they noticed it.

There are some kinds of businesses where making sure the regulators like what you’re doing is pretty much a prerequisite. On the other hand, plenty of companies got where they are today by pushing the limits.

[1] https://www.ftc.gov/business-guidance/blog/2023/03/chatbots-...

Wow, this is a great article. Obviously writing is easier than enforcing, but I'm pretty impressed with whoever at the FTC is already thinking so clearly about this stuff.
Thanks for sharing this ^
(apologies for going off topic here)

Wow. I would have imagined an article from the FTC to be more... Bland, for want of a better term.

The FTC consistently has one of the absolute best author voices in all of government. Pick a blog post at random and see what I mean. Their index on tech is probably the area you have the most domain knowledge in and so it’s probably the best area to evaluate them: https://www.ftc.gov/business-guidance/blog/term/1428

Clear, direct, confident, not overloaded with qualifiers, not afraid of metaphor, self-summarizing, signposting, and most importantly it always has an energy of some kind that government communication (in seeking to appear neutral) regularly lacks - having that energy is why it doesn’t feel “bland”. I wonder if they have internal documents to guide their writers, or if it’s mostly information stored in the heads of Lesley Fair and Michael Atleson (who between them seem write most - all? - of the posts).

[flagged]
Wouldn't fewer people having access make your proposed scenario less likely, regardless?

If other people continue to have access as well as the '3 letter agencies', the same power will still exists for the agencies, except that there will also be an essentially unlimited number of other people who could be used as scapegoats.

If only '3 letter agencies' have access, they would obviously be the first ones to come under scrutiny if a case of misuse were discovered.

Yes but people will continue generating meme recordings like the ones going around showing the recent POTUSs gaming dialogue. Thus showing everyone not to trust anything.

Without that, we'll just never know, and Joe Blow who never saw a deepfake of Joe Biden praising the stickiness of the latest OG Kush will trust anything.

I'm with you on this. I can't honestly think a good use case for the average user to generate audio this way. Maybe some niche use case in like movie or tv production where you can generate a missing line without flying in an actor or something. Or maybe for generating dialogues for videogames. But those are business use cases, not things for the genral public.
Pretty much any hobby video game development or animation?
So basically stealing other people voice for you hobby? Great!
Who says it has to be someone else's voice?
I mean, if it is your voice, can’t you just record it directly? Isn’t that better than going through an artificial middle man that has to be primed first with some recording of yours?
It would be a huge time saver. Typically a pro does around 100 lines per hour, an amateur doing multiple takes would be significantly slower. So a character with 1000 lines could easily be 20-30 hours of work, just for a first draft. It would be pretty amazing to be able to just revise the script and auto-generate a new recording, even if the quality is only 90% there.

Just like with image generation models, this will massively raise the bar for what amateurs can do with a limited budget and limited time. It's hard to justify spending thousands of dollars on voice acting and art for a hobby project, but now amateurs can get something that is 90% there and substitute professional work if the project takes off.

I'm 100% certain that people will offer up their own voices as open-source. I would be happy to, albeit maybe anonymously.
Artificially generated but faithful to the original voice for people losing theirs or being unable to speak for various medical reasons. Obviously the fun/deceiving use-cases are much more numerous.
When my sons were young, I would tell them elaborate stories where they were the main characters. I recorded most of the stories, but it is full of verbal fillers (um,ahh), since I was making it up as I went. I would love to convert the audio to text with Whisper, filter out fillers and then output the cleaned up version in my own voice. I could see this type of workflow being very popular with podcasters.
You should absolutely do this, but please skip the Whisper.

The reason is that if you speak with lots of verbal fillers, that's actually an important part of how you sound to other people. It makes sense to clean up audio for a podcast, but not for your great grandchildren.

A voice cloner doesn't care that you say "um" too much. It's parsing audio for phonemes.

You can already do what you are aiming for, without the transcription part (removing fillers with word-based filters does not give great results compared to removing with voice-based filters).

You can use Descript, CleanVoice, and other tools to achieve exactly what you just said, in a few minutes, from just the original recording.

Personally, I would find this very useful. I (used to) create internal tech training videos for our organization and would routinely stumble when doing voice overlay. Even though everything was scripted out, it took lots of editing time to get the audio and video aligned without the vocal stumbling (ahs, ehs, silence, voice redos, etc). Just my $0.02
My Dad died before my kids were born. When he got cancer, he recorded himself reading "The Night Before Christmas", which is about enough audio for the high quality version of this technology. Is it ghoulish on my part to want to hear his voice again or for my kids to hear it? Maybe. Do I really care what you think (or really what _I_ think) about that? No.
Sorry to hear. Hope your kids enjoy his recording.

But yes, it would be weird to generate more stuff spoken by your father by using this technology. And beyond that, what's even the point? It's not your dad.

It doesn't really sound healthy to generate content of loved ones.

Yes in a few years you would be able to generate a complete avatar of someone, but it isn't them, and i think it will mess with you mentally.

People who would tell you not to use your recorded audio to create more simulations of your father speaking are the same sort of folks with strong opinions about what other people do in the bedroom.

I happen to be someone who believes that it's wonderful your dad left you with this artifact. It was a touching sentiment then, and now it can serve his obvious purpose many times over.

He didn't record himself as a side-effect of disease, or because he loved that particular story in the sound of his voice. He wanted people in the future to be able to hear what he sounded like!

Given that he could not have foreseen voice cloning (and therefore not explicitly asked for it) I cannot think of a more obvious example of someone wanting their voice to survive them.

I wish more folks would record The Night Before Christmas.

You shouldn’t care what I think. You shouldn’t care what anyone here thinks. Creating fake memories is not something I’d ever consider doing but that’s just me.
We're getting to the point where all voice conversations will need to be authenticated via OTP, even between family members, on the phone. Especially for banking, etc.
Voice conversations on FaceTime, WhatsApp, etc. are already authenticated. Perhaps it's time to stop using non-VoIP calls?
Can you call your bank with that?
I wish I could. But unlike my family members, the bank doesn't authenticate me by my voice.
It is a big issue in India. We have a few Bollywood celebrities with "trademark voices" - voices so distinct you would instantly associate it with that celeb. There is a Huge mimicry culture with 100s of extremely talented mimics who can clone any voice. On top of which, there is a gigantic radio audience, so the celebs despite making millions in Bollywood films, advertise cement, coconut oil, fountain pens, tobacco, beauty creams, online casinos etc in radio clips, using their distinctive voice.

This makes for a rather explosive combination. I could, as some tobacco exec, hire some mimic to promote cigarette sales using a celebrity's distinct voice. By the time the regulators catch up, the spot has aired a few million times & made a potload of money.

A bunch of celebs[1][2] have trademarked their voice...but enforcement is spotty.

[1] https://economictimes.indiatimes.com/news/new-updates/amitab... [2] https://www.financialexpress.com/archive/when-celebrities-se...

Do you want to tell us why it should be illegal? Comparing something like this to nuclear weapons is a bit hyperbolic, at least without giving more context.
Nuclear weapons are only ilegal if you don't have them.
Serious question. What is the difference in implications between this and a professional voice impersonator? I don't think it's as dangerous as we think it is. All of the consequences that Play.ht bring to society are already possible today and have been for some time. The difference is that it will be easier, but I don't think that makes it any more dangerous.
there's a massive higher bar in effort and costs in getting an impersonator

whilst this is cheap and easy - increasing the potential for scams in big way - even to the point of automating the scam

> The difference is that it will be easier Seems like you know what the difference is, you just haven't assigned it the proper weight.
Two differences I can think of, either side the debate:

1) The impersonation can be carried out in real-time by the criminal themselves. No need to employ anyone else. (No trail leading to them.)

2) Pro impersonators aren't common in society. They are limited as an asset and not duplicatable. So, using one cannot spread like wildfire and overwhelm our awareness that voice impersonation is something of a common risk.

Maybe the second could hold the first in check. I think disruptive tech like this & similar advances in visuals come with a societal impact that lessens potentials for realising the bigger fears. But people just love fears.

Scaling + ease of use.

Compare:

1) Use lots of time to find person who can impersonate specific other person. Unless you give them a lot of money or threats to shut up you can't use them to real time decieve someone.

2) Clone 1 million voices from tiktok in 1 minute. Contact 10 million relatives with a synth voice that is intelligent enough to answer questions.

We will have billions of AI's, containers, programs, agents running around trying to deceive absolutely everyone and their grandmother 24/7 soon.

Put the recent tech advances together: we can now ask bots to generate an identity online that looks like a legit human (with pictures, audio, text).

Of course a human could do that manually, but with those AIs it's a completely different scale, and it can be automated (so someone with no skills can click a button "generate 10k fake identities online").

Maybe even with one click, those techs could generate a fake coworker and send phishing e-mails. Suddenly every single e-mail you receive (or friend request or call) could be a very convincing fake. You don't have to be a high-value target anymore, it's all automated.

That makes it much more dangerous: from "can be forged manually with time and resources" to "everyone can do it at scale for free".

It's amazing how many think declaring something illegal will stop criminally-minded people having and using it.
It won't stop it, but will allow enforcement agencies to enforce. Otherwise, they have no legal recourse to do so.
Enforce what?

Why should the ability to impersonate a persons voice suddenly become a crime in itself?

Should we arrest Jim Carrey?

Isn't it when the thing was used to do something else illegal when enforcement is required?

Sure but they can make extra penalties for using these in illegal acts. For example robbing someone, and robbing someone while using a gun get different sentences.
It's already illegal to defraud someone into sending you money.
I have a "large" (~40K 10s lines) corpus of captioned dialogue from a video game that I briefly investigated training a model similar to this to "clone voices" with, but pretty quickly came to the realisation that doing so would be pretty unethetical to all involved.

It became more apparent to me how icky this is as the voice actor of one of the most iconic characters in the game died suddenly 10 days ago...

There’s a repository with a VITS model for English/Chinese/Japanese voice cloning that was pretrained on Genshin Impact voices.

I’m not actually sure whether that’s illegal (in the civil/IP sense), provided you’re not using the original voices to synthesise text.

Right on the top of their page is the example: "good afternoon sir, I will just need your credit card number and security code to proceed." Wow.
> This is the first startup here where I think the tech should essentially be illegal.

Yes, I agree, only criminals should be allowed to freely run it.

You are right, the technology will become ubiquitous, therefore, at least for platforms like us, it's a responsibility to have countermeasures and safeguards to prevent abuse and harm people. There'll always be people who will find ways to abuse but making it more and more difficult and evolving on that seems like a way forward.

We have these measures in place and are working on others to make sure the technology is used towards the betterment of humanity.

1/ Auto moderation on text to block harmful/malicious speech. 2/ As someone pointed out in the comments, we had a manual review process in place where the user is required to read out a consent and a member from Play.ht would review it before approving the voice. We're working on improving and adding this back. 3/ The user facing service is paywalled so we don't allow everyone in. 4/ Users trying to create malicious content are flagged and reviewed. 5/ A classifier to detect AI generated speech

Cofounder here, What you see in the above demo is a very rate-limited demo of our upcoming model. We realize how dangerous this technology can be and have built a lot of mitigations on our main product (Play.ht) to reduce possible abuse: - We strictly moderate the generated text of any sexual, offensive, racist, or threatening content. It automatically gets detected and blocked.

- We built and are offering for free a tool that can identify AI generated vs human-generated audio (https://play.ht/voice-classifier-detect-ai-voices/), we will continue to invest in this tool, and we hope it helps with deploying this technology safely.

- If we get any reports of a cloned voice without consent, we block the user and remove the voice instantly.

- The price of high-fidelity voice cloning is too high for scammers to use at scale; we have been live with it for four months and haven't had any cases of abuse so far.

Like any technology, it has the potential to be abused, and we are working hard to mitigate that and deploy it safely. We will continue to observe the use cases and user feedback and improve the safety of the service accordingly.

Since we launched voice cloning 4 months ago, we have seen enough genuine use cases which motivated us to keep moving forward and figure out safe ways to make the technology useful for all.

> haven't had any cases of abuse so far.

How do you know this?

>We strictly moderate the generated text of any sexual, offensive, racist, or threatening content.

This won't be the problem. My voice calling my parents asking for money to be sent to a random account will be the problem. And none of that will be sexual, offensive, racist, or threatening.

>we are working hard to mitigate that and deploy it safely.

How?

>we have seen enough genuine use cases

What?

Exactly. Would love to see some testimonials on that...
> We strictly moderate the generated text of any sexual, offensive, racist, or threatening content.

This is exactly what makes me so angry about "AI safety" initiatives: they are largely worrying about the wrong thing. People have been so focused on the "this may make some obscene joke, or be biased against some skin colors" that they have completely missed out on the much more serious harms that AI will cause with respect to, in this case, impersonation scams.

Congrats, people can't say the N-word with your technology, but they can say "Hi Bob, just calling to verify that we did indeed change the target account where you should wire your invoice payment."

Honestly I'm starting to wonder this about AI in general. I mean realistically, there's a decent chance will be looking at general AI soon. The best-case scenario endgame of that is creating a benevolent God. It might be time to start asking ourselves if that's what we want.
Pretty sure that's something that ought to have been discussed before any of this ever started, but you know, scientists, could, should. I look forward to the chaos and destruction and all these "brilliant" software developers wringing their hands saying they couldn't possibly have imagined such horrible outcomes from their fun money-making venture that just so happened to undermine the concept of a shared reality.
I think we all thought we'd be able to come to those decisions on a more gradual timeline. The breakneck pace of AI breakthroughs over the past few years have revealed: not so much.
I find myself doubting that "we all" would have been able to have such discussions or ever make these decisions ourselves. Silicon Valley and big tech spent the last few decades hijacking human psychology and employing dark patterns in technology that was supposed to be "democratizing" and "empowering" in order to maximize profit. Now we stand at this precipice, coupled with the RESTRICT Act, which I have no doubt will pass.

All's well that ends well, though. We simply don't have the resources to continue this "breakneck pace".

How is the latency for real-time TTS? I remember kicking the tires several months back but went with one of the big 3 cloud providers since they had lower latency.

I also like that the cloud provider supports SSML and I can explicitly configure the emotion, whereas Playht dynamically changed the emotion based on context of the text.

Low latency would open up a whole lot of interesting applications. Even Elevenlabs doesn't seem to have low enough latency in my testing to work as a convincing voice assistant or to, for example, work in real time on a phone call. For that we likely need QUIC or some kind of streaming protocol.
The latency is not real-time yet but we're working on getting it to near real time. Regarding controlling the voice, we've added a few params like rate, voice guidance and temperature but for the most part the emotion is dependent on the text for now.
YC is assisting / help to fund the next big telephone scam in which more and more people across the world will fall victim to cloned voice audio scams. Grandma is called by her grandsons voice asking for money, but it's not him. YEAH!

I had envisioned the scammers leveling up to this scam last week or so in a comment here. Though google news a few days later showed me it's already happening...

https://www.dailymail.co.uk/news/article-11897239/Houston-co...

This is a good reminder that we all need to have a "safe word" that we can use to verify to the important people in our life that the voice they may be hearing on the phone or elsewhere is really us.

Get a panicky call from "me" in the middle of the night? If I don't include my safe word, that call isn't from me.

Very good suggestion
That scam was popular here in Argentina a few years ago. We call it "virtual kidnaping" https://www.fbi.gov/news/stories/virtual-kidnapping , nobody is kidnaped, it's just a scam using a phone call.

It's not very important that the voice is similar to the supposed victim. Usually the person in the call is weeping and it's very difficult to recognize the voice. Moreover a confusing voice at 2am may be interpreted as any of your relatives or friends, but an exact voice can be interpreted only as one and it's easier to know that that person is safe.

Some scammers tried to pull this scam off on my stepfather years ago. He got a call that, through the wailing and tears, told him that I'd been thrown into a Mexican prison and needed bail money immediately.

He was 90% convinced that it was true, but my mother made him call me before doing anything, which saved him about $10k. She thought it was suspicious that I would have left the country without mentioning it to her.

If the person he was talking to was relatively calm and sounded like me, it might have been successful.

At least here, most were just calling random telephone and letting people guess who is the kidnaped person

> Bububu. Hi, I'm ... bububu

> John?

> Bububu. Yes, I'm John. bububu. I'm in bububu ... the jail in ... bububu

> Mexico?

> Bububu. Yes, In Mexico. bububu. And I need money ... bububu

There are other that research the victim and have more data for a targeted call, but it's more difficult so most cases where random calls where they don't have a sample voice of the victim.

It is frightening that we have gotten to this point already.
Society definitely needs to adapt to this new norm; we are trying to roll this out as safely as possible, but others are not as careful, and this technology will just become more ubiquitous over time.
Wow the actual speech part is terrible, the number of mispronunciations is surprising.
Is there a way for me to preemptively request (demand) my voice (likeness) never be used by this service? How would one go about doing that?
Looking at the voices they use as a demo on their blog post (https://play.ht/blog/introducing-truly-realistic-text-to-spe... is just one example) I don't think consent is really on their radar.

Their FAQ says

> Can I clone anyone's voice?

> Yes, we allow you to clone another person's voice if you have their consent. As you can imagine, cloning a voice which sounds exactly like the person is a powerful thing and can be easily misused. We deeply care about ethics and privacy and have implemented verfication processes and regulations to avoid people cloning anyone's voice without their consent.

But I very much doubt that they've gotten the consent of even half the celebrity voices they're using to promote their service.

Hey HN, we are Mahmoud and Hammad

Are you though? You might just be computer-generated.

While I'm very impressed with this technically (and as a pro-audio person I feel validated to see my predictions of a few years back coming true so dramatically), I don't see anything about risk management in here. Your tech absolutely will get used by scammers, given the overabundance of voice data on the open internet. How are you going to hedge against that?

We have many mitigations in place to increase the safety of this service, I mentioned some of that here https://news.ycombinator.com/item?id=35331310
That's interesting. But I think it's a mistake to focus on relying on price to prevent abuse at scale. The use case for abuse of this technology is in highly targeted frauds, not broad-spectrum scams like insurance robocalls. Additionally, this will be zero deterrent to deep-pocketed actors like political action committees that generate fakery to influence elections and the like.

I'm trying not to be reflexively dismissive, and I know the technology is evolving so fast that your individual company can't necessarily pre-empt it, any more than an email software supplier is responsible for the existence of phishing. But I work adjacent to the security space (studying violent extremists) and I can think of a ton of ways to abuse this where economics would be absolutely zero deterrent.

Wow, I haven't even thought of that. Imagine this being used together with a chatgpt equivalent. Scam rates are going to go through the roof.
(comment deleted)
Wow hammad, here on HN :P couldn't believe until few years ago
When OpenAI and Google insists on "safety" it leaves the door open for startups that do things like this.
Seems like we'll sort of inevitably end up with a sort of cross-site verifiable identity on the internet. All content requiring some sort of verified user backing it. Generally will be interesting to see what an internet with less anonymity looks like.
> Generally will be interesting to see what an internet with less anonymity looks like.

It will be much more dangerous, in all probability.

> We are thrilled to be sharing our new model

really? that's a nice gesture… so where can I download it?

I'm having a hard time coming up with a non-nefarious use case for this.
Voice generator tech has created some decent surreal memes (like audio recordings of Biden, Obama, and Trump playing video games together).

Outside of memes or maybe the occasional well-intentioned prank, I really can't think of anything either.

Massively reducing costs for Voice Over in Video Games. This should make it even feasible to create mods with audio which would be great :)
I think “talking” with dead relatives or friends will become real pretty soon.

If people can find comfort hearing their mom say words of encouragement in a tough situation, I think a lot of people would do it. Kinda hard because for some others that would mean never getting closure.

Weird stuff is certainly about to happen…

The last thing on earth I'd want is for any aspect of my dead relatives to be reanimated through technology. No. That's absolutely fucking horrific to consider. I don't need a hallucinating AI pretending to be my dead wife. That's literally shambolic.

There is vastly more potential for that to be abused by others than used in any emotionally or socially constructive way.

I would also find that very creepy and it would probably keep you from moving on. I think there is a big difference between remembering what happened by looking at a photo or hearing an audio recording and having newly generated "content" from a deceased loved one.
there has been some media coverage on this already (e.g. [1]). an emerging concern among mental healthcare professionals is that a sufficiently-convincing simulation could interfere with the progression of the stages of grief, prolonging the 'denial' stage and potentially heightening the intensity of the stages that follow.

[1] https://www.wired.com/story/a-sons-race-to-give-his-dying-fa...

I would consider studios taking voice actors' voices and using them to generate new content beyond their contract to be abuse. I'm sure big corporations are rubbing their hands in anticipation, but I'm sure killing the VA industry will make the world just a tiny bit worse for everyone else.

Mods are more difficult to attach a moral judgement to. I don't think I'd really consider them malicious, as long as they're not sold, but there's a very thin line between a high quality mod and stealing someone's voice.

On the other hand, why shouldn't voice actors benefit from this tech?

I can easily imagine a future where AI-generated impersonations are deemed by courts or new legislation to be protected by personality rights. In that world, voice actors could expand their business by offering deeply discounted rates for AI-generated work.

Alternatively, if/when tech like Play.ht is consistently good enough, maybe it just becomes a standard practice for all voice acting work to include a combination of human- and AI-generated content, like a programmer using Copilot or a writer using GPT.

I'm sure programmers would love to expand their business opportunities by offering deeply discounted rates for creating AI-generated code.

No? Then why do you assume that someone else would want to do the same in their profession?

As AI-generated content is not protectable under IP law, it's a non-starter for games, film, TV, or music for anything except background filler.

(comment deleted)
Sure, why not? If you could earn more money and produce more value to society with the same amount of labor, and the legal/regulatory environment supported it, I wouldn't see a reason not to.

If you had a solo contracting business, and the technology existed to fully outsource a development project to AI based on carefully documented requirements, using it would be a cheaper alternative to subcontracting. Rather than writing every line of code by hand, you would transition to becoming an architect, project manager, code reviewer, and QA tester. Now you're one person with the resources and earning potential of an entire development shop.

I have my fair share of complaints about AI coding tools, but that isn't one of them. Maybe the increase in supply would result in a lower average software engineering income, but it wouldn't have to if demand kept pace with supply.

Furthermore, code is more fungible than a person's voice. If someone wants a particular celebrity's voice, that celebrity has a monopoly on it. Thus, it's not obvious that increasing the supply of one's voice acting work would decrease its value. (I suspect the opposite to be the case, until a point of diminishing returns.)

Although the voice acting case has a similar concern; will we get an explosion in new and/or higher-quality media, or will we see a consolidation to a smaller number of well known voice actors taking an outsized amount amount of work? Another issue, if we look beyond impersonation specifically, is that human voices may become marginalized over time in favor of entirely synthetic voices. I imagine that this would start with synthetic voices playing minor roles alongside human/human-impersonated voices, but over time certain synthetic voices would organically become recognizable in their own rights.

Again, I see plenty of concerns with AI in general, but more of a mixed bag than strictly negative, and there isn't anything inherently nefarious about this product in particular.

Personally, I'm optimistic about what society looks like in the long run if humanity proves to be a responsible steward of increasingly advanced AI. By the time we're at a point where 90% of people can be effectively automated out of a job, we'll have had to have figured out some alternative way of distributing resources among the population, i.e. a meaningful UBI backed by continued growth of our species' collective wealth and productivity. I can easily imagine a not-too-distant world that is effectively post-scarcity, where it's not frowned upon to spend years (or lifetimes) on non-income-generating pursuits, and where the only jobs performed by humans are entrepreneur, executive, politician, judge, general, teacher, and other things of that must be done by humans for one reason or another.

So am I happy that AI is encroaching on skilled labor? In the short term, not necessarily. But it's not necessarily bad either, it's the reality that we're in, and long-term I'm more optimistic than not.

I think it will probably kill the current Business Model of the VA Industry. Having the ability to generate as much audio content as you like without the risk of the VA not being available anymore (dead, booked out,...) is just too good to pass up.

Instead we will probably see licenses for generated voices. And in case for games the game developer could make the voice model freely available for mods of his game.(The mods are already using assets from the game, why not also audio?)

Machine generated content cannot be copyrighted so I doubt companies will switch to AI generated voices for big games for that reason.

Voices can't be copyrighted either, so I don't see how a license for a generated voice would even work.

Star Trek: Prodigy has already used audio from previous movies and TV to bring back to life several actors from previous series. It's not exactly the same as this, but their dialogue was taken out of context to create new scenes and story.
I know, and I almost wished they did use AI for that segment because it was pretty jarring (especially the TOS recordings).

There's still a huge difference between "reusing the work the studio paid for" and "recreating your voice forever after doing a single project".

Anything written can be listened to with this tech. Any news article, any short story, a draft of a piece of writing you're working on. There is too much text for human beings to read it all.
> There is too much text for human beings to read it all.

so your logic is that all that text should be audio and people will consume more? Because I got news for you, reading is faster than listening.

Oh yeah, how does reading work out for you while you're driving a car? smh...
When I said there's too much text for human beings to read it all I meant that it isn't feasible to pay people to read all text that someone might want to listen to into audio. Like a random blog written by someone in their spare time probably isn't going to hire a voice actor.

I think the case for having all text be listenable is pretty clear. We're all really busy and often our hands are busy but we're not doing something that mentally stimulating. This is an ideal time to listen to an audiobook, a blog, the news, or whatever else you'd like.

And all AI bots are here to generate even more text. :( We will need to rethink and reevaluate lots of things that we are used to.
I'd get a kick out of having my own blog posts read to me in James Earl Jones's voice.

Or, heck, my own voice. Though it'd be surreal to hear not-me-but-me saying things I've never said.

even this is ethically questionable. james earl jones's voice is his livelihood.
While that is true, I'm not suggesting a pattern of behavior - just that it would be fun to hear.
Generating audio for an audio book: If an author could speak for 20 minutes and then generate audio for an entire book from the book's text and the model, I think that would be very useful.
20 seconds*
The OP mentioned that for so called, "High-fidelity voice cloning", it would take 20 minutes of training. I think a book author would want the best quality possible to reproduce their voice.
Why reproduce their voice? There's no value-add there.
Many people prefer an audiobook version of a book to be read by the original author, which isn't always the case. If an author could make that version happen by using 20 minutes of their time + text2speech of the whole book, that would be an immensely positive value proposition on the side of this company.

But I'm not sure. Part of why I'd prefer the original author to read a book is that they vocally emphasize certain parts of the book, and I don't think these models could do that at this point.

> Many people prefer an audiobook version of a book to be read by the original author

Right, but having AI read the book in the author's voice is definitely not the author reading the work.

As you mention, the reason that people like to hear the author read it is because it's the author reading it, theoretically emphasizing and acting things out according to what was intended. It's not just to hear the author's voice.

So I don't see what the value-add is.

I am toying about with building a virtual puppet software in the style of watchmeforever. I have a number of voices I do for the stage and DnD that I would be willing to train a few models on so I could give my puppets unique voices.
We have been seeing some of these genuine use cases: youtube creators, audiobooks, elearning videos, podcasts, commercials, dubbing, and gaming.
BS. That could just be done without imitating someone's voice.
No-one is going to listen to an audiobook made with this. It's still fundamentally just TTS.
Have you tried it? I've listened to 2 generated audiobooks so far, has been great
I'm using this kind of technology for temporary voice tracks in animated shorts.

I'd really like something like Img2Img for voices so I can translate a performance to an arbitrary (synthetic) voice.

Tortoise TTS can do this. You just pass it your example as a conditioning latent.
Listening to the demos I'm not entirely convinced by this (https://playground.play.ht/listen/189 was pretty funny). I wonder if this company will end up taking down (and subsequently pricing out most people using this tech for fun) arbitrary voice generation just like its competitors have so far.

Going to the demo page and hearing a random snippet of Musk-worship was pretty weird. Out of all audio tracks to place at the top of your demos, you chose this?

Wow, I call to the team behind this. I really STRONGLY think you should at least implement some sort of URL stealthing. I'm not a Web Security expert, but it reminds me of a talk where some company just made medical records 'public' like this.
On the contrary, this should be accessible so we can see what people are generating.
I see a bright future for play.ht in the "pre-event" audiolog generation market. Somebody get ubisoft on the phone.
AI can now generate youtube poops