This is a bit different. These audio clips use the default voice of each of these systems. I was asking about zero-shot voice cloning, i.e. transferring a recorded voice and synthesizing speech in that voice.
I tried zero-shot voice cloning in all of the top OSS models in the Arena and performance was bad.
Thanks for the flag. VoiceCraft is indeed the best ZS OSS voice cloning tool, despite appearing at the bottom of the TTS arena They have a really easy-to-use gradio demo on their repo if anyone else wants to give it a try.
There is still a big gap between 11Labs and Character.ai and the VoiceCraft voices would not be confused for the real speaker, but this is much closer.
Healthcare is a slippery slope. Don’t tax hospitals, but tax pharmaceutical companies? Don’t tax pharmaceutical companies, but tax the income of doctors?
I think the point of the article is that it’s really hard to make these judgments and doing way with the whole system entirely makes the most sense, which seems reasonable to me.
"Best I can do is tax you like you have socialized medicine[0], then force you to pay 2 - 4 fold what those other counties private systems cost to actually get care for most of your life." - Every politician in my entire life time
Quite a few politicians suggest we take one of the various approaches the rest of the OECD countries do, as they're clearly more cost-efficient... but that gets derided as socialism.
They also never make the point that we are already paying for it. They want to tax me more, when if they were as efficient as the NHS I'd have two healthcare systems with money left over.
Medicare for all will never see the floor until everyone currently over 60 is gone, at a minimum. They are all in on it.
The US can clearly afford both, as your chart demonstrates. It shows we pay more in tax than the other OECD countries pay in total, including taxes for healthcare.
We've picked the least efficient worst-of-both-worlds scenario out of a fear of anything that sounds like socialism ("public option", "Medicare for All", etc.).
Its corruption. Explain it in the terms I have on the national debate stage, everyone is on board. NHS style healthcare AND lower taxes. You'd win every election, but you'll never see that stage.
> The US can clearly afford both, as your chart demonstrates.
The chart doesn't demonstrate anything of the sort. By definition, sovereign governments can afford literally anything, since they can print the currency within the country. It will just cause inflation.
That statement is complete and utter reductionist nonsense.
The US government is currently paying as much per capita for healthcare as socialized systems, then individuals pay again. If we can afford what we're doing now, it's obvious that we could afford a socialized system. It costs less.
The reason we don't have one is because Democratic politicians, funded by healthcare and drugs, are so willing to lie about the math for money, and the media, funded by ads for healthcare and drugs, are so willing to help them.
The fact that our current setup costs 2-3x any other OECD nation indicates reform to be more like their systems is likely to reduce costs, not increase them.
> Don’t tax hospitals, but tax pharmaceutical companies?
Hospitals, to maintain their nonprofit status, have to provide charitable contributions. One way is when patients apply for payment reduction or discharge because they can not afford to pay. Patients need to apply for this so they have documentation to give the IRS.
> Don’t tax pharmaceutical companies, but tax the income of doctors?
How often do you ask a doctor to work for free? They still get paid by the hospital.
Yes, pharmaceutical companies, which as far as I know are private, also have something like this. This is preventing media coverage and Congress imposing max cap on prices on public assistance programs (prices private insurance will use for leverage) so they can keep charging what they want.
> I think the point of the article is that it’s really hard to make these judgments and doing way with the whole system entirely makes the most sense, which seems reasonable to me.
The point of this article is to attack the nonprofit competitors to their due paying members. They gloss over many details and as I posted elsewhere; they even defensive of the tax loopholes of their due paying members.
> How often do you ask a doctor to work for free? They still get paid by the hospital.
Oftentimes, physicians are not paid by the hospital at all. They will form consulting groups and work for the consulting group which will have a contract with the hospital, or they'll do something C2C.
> Oftentimes, physicians are not paid by the hospital at all. They will form consulting groups and work for the consulting group which will have a contract with the hospital, or they'll do something C2C.
This is a red herring. The hospital still pays for a service (doctors or a group representing them) while being required to make charitable contributions to keep it's nonprofit status. These charitable contributions are very important for people in poorer and rural areas of the United States that for profit entities do not want to operate in (or would demand subsidies from the government to operate in).
I thought that this earnings announcement would be the first one where we'd see some impact from competitive knowledge engines, e.g. Perplexity, You.com, ChatGPT + Bing, etc., but Google still grew search $6B/15%.
This is impressive both because it's hard to keep such a big business growing at that rate and because essentially everyone in my social circle has moved on from going to Google first for information. I guess our demographic is not predictive of the larger market.
I think that most of the traffic being stolen away is going to be for low value searches. I (and probably almost everyone else) use Google when I already know what I want, ie I’m trying to get to a company’s website to buy a particular thing, but don’t know their url name.
I’m not going to use an LLM to shop for car insurance or look for hotels.
Ok, but if you’re a well off 40 something mom trying to buy your daughter tickets to a Taylor Swift concert, what would you do? I’m willing to bet that the proportion of people using Perplexity to those using Google to search for concert tickets this past quarter was infinitesimal. Or those shopping for new cars. Or airline tickets. Or furniture. Or lawyers. Or health care. I can go on.
>I’m not going to use an LLM to shop for car insurance or look for hotels.
As I understand it, this is a great use-case for AI agents (a-la Custom GPT) and I wouldn't be surprised if the tech for that matures over the next year.
I guess I could imagine how they might be helpful in initial research - but no way I'm letting some LLM book a hotel that might not exist or get me car insurance from StratesFrarm Unsurance.
I'm not clear how this is different from scammers targeting humans? If anything, I'd expect it to be easier to set up the LLM agent to have secure/cryptographical validation of the providers, to prevent against the sorts of traps that humans fall into.
In the simplest case, we could just have trusted authorities provide lists of verified booking partners, same as e.g. Google Flights does now.
The whole agent idea is questionable anyway. Sites are are already optimized to require the absolute minimum effort for conversion and most of the interesting aggregator sites already exist and are similarly optimized. Who's going to want a clunky natural language interface that breaks half the time?
I would.
I don't know what you're referring to as "absolute minimum", but when I book a flight I'm typically taken through a flow of about 10 pages, each trying to convince me to buy another set of extras. I would love an agent whom I could tell to not buy any extras and just get the basic plane tickets for me in one click, or if it fails to do so, to try again with another partner.
That would be the first step of rollout. The next step would be to ask you in ways that make it hard for you to say no. An LLM agent isn't going to stop companies from bad behavior/dark patterns, it's going to make them more effective at deploying dark patterns to segments of the market.
I think only competition and/or regulation helps push down the allure that dark patterns present to companies.
Will local AI agents be the "linux desktop" of the next generation? Because as I understand it, next year is gunna be the year that local AI agents become a thing for the normies.
For the average desktop user, Linux never had anything to offer over Windows. The last nail in the coffin was obvious to me at the time: Linux didn't make any headway when Microsoft started shoving the first ridiculously-bad versions of Windows 10 down everybody's throat by abusing a security update channel. At that point it became clear that there was simply no demand for Linux in the desktop market except at a niche hobbyist level. Stockholm Syndrome was in full effect among Windows users, and it still is, and, well, I guess that's just the way it goes.
The enshittification of the Internet is even worse, though. I think personalized search and content curation through the use of local AI agents may be the only way to buck that trend.
I want something that will do for companies like Google, Amazon, and the news media what SpamBayes did for email: make it useful again. Local filtering by a tireless and incorruptible proxy whose interests are aligned with my own is the only way forward IMHO. The only question is how long it will take. My rather-useless guess is more than a year, less than ten years.
Google Search is a product embedded in the lives of everyone on the planet. I wonder to what extent its growth is a reflection of increasing population and Internet penetration. Even if "competitive knowledge engines" impacted Google Search, the effect would likely be minuscule compared to existing growth trajectories.
Speaking personally, there are some queries where I prefer an LLM. But usually I start with Google, and it's only after 5-10 searches that I get frustrated enough to remember I could just ask ChatGPT instead. So ironically, I actually send more searches to Google than I would have if they gave me the answer on the first one.
I wonder what their search metrics would look like if they removed quick bursts of searches. Presumably, someone searching five times in a row is actually having a bad experience, rather than loving the product so much they came back to it five times in one minute.
I think this is the same thing I feel when I read Meta's earnings. I continue to see growth in usage of their core properties but I don't think my entire circle touches Facebook or even Instagram nearly as much.
Whatsapp certainly isn't driving those ad dollars so it truly is remarkable how disconnected my demographic (loosely using my here) is from the overall world usage.
You have no idea of Whatsapp's marketshare in India - all small businesses use Whatsapp for all communication and even take payments. That's a 1.4B population.
Maybe not but they have all the timing info, the identities of the chat participants, and the facebook surveillance network. They also used to leak URLs pasted into the app out in order to generate previews, so they know what you’re linking to from within the app. Us they can infer way more about you than seems obvious at first.
They are encrypted to your device and your device can upload them unencrypted in backups if you don't configure encryption. Plus the original argument was Meta did not have access.
It's not about money - how valuable the company is depends on how many people use it intersect how useful it is.
As long as it's valuable, it's valuable.
I have a great idea of Whatsapp's marketshare globally. I am simply pointing out the ad revenue of Whatsapp is extremely low compared to their other properties. There is simply not enough surface area in Whatsapp used for that and their business profiles etc. are not going to be outearning US users of FB/Instagram.
Ultimately advertisers want a return on their spend and it's a closely watched metric for marketers. Them continuing to spend is indicative that there's growing value in ads, ie bots/ai agents cannot be the reason for their growth.
You could argue fb and particularly twitter are incentivized to include it in their DAU counts but market cares more for revenue for large companies.
SMEs spend a lot on advertising, in aggregate. These smaller businesses are more sensitive to lowered conversation rates and will bail early - Meta took a huge hit in earnings in the aftermath of Apple's privacy changes due to poor conversions. Advertisers didn't keep pumping money in - they stopped campaigns.
So do Android and iOS. They're still the primary indicators of the strength of the smartphone market. You probably can't imagine a world without advertisements or smartphones, but that's the point from the post you're responding to that you're missing. If the advertising industry looks healthy, it helps everyone by keeping investors satisfied and share prices up, even if it's a ruse. It is possible for smartphones and the concept of advertising to cease existing some day. There's a spectrum between the current reality and absolute zero, and staying as far away from zero is the goal, even once we enter an inescapable decline.
Arguably true, but that's a rather different point than "there is no alternative". There are arguments to be made about market "health" and the relative sizes of the largest players, but what the upthread comment was doing was resorting to "anti trust" shorthand that clearly doesn't apply. There may be problems with this market, but lack of competition isn't one of them.
> essentially everyone in my social circle has moved on from going to Google first for information. I guess our demographic is not predictive of the larger market.
"Google is dead, no one goes there anymore" is one of the most tired takes I see frequently on HN. It's nice to hear someone express the self-awareness to realize that what they see in their immediate circle is not representative of reality.
They think if they say it enough times they’ll eventually get it right one day to have their I told you so moment.
The smart money is always going against their hate. It’s not enough to stop using their services, but to also convince their friends and family to switch to their choices.
I'll admit that despite having a GPT4 sub which I use often, I still fallback on google for quick questions and verification of what GPT4 says if I need to be certain.
Not even because I am particularly going to google, it's just so heavily integrated.
I find GPT4 requires long queries to get much out. And I can't be bothered typing out a long query if Google can give me the same in 3 words and scanning down the result page.
I have moved to Kagi as my search engine, although 99.99..% of people will probably never pay for search. Anecdotally, whenever I'm not finding what I need on Kagi and fallback to Google (less than once per week), it's shocking what a disaster the UX of Google's products have become (Maps is an even bigger offender than Search, where seemingly arbitrarily labeled places get more visibility than the search results).
I don't hate on Google, but the user highlighted some feature on Google Maps that are not desirable, and clearly knows more about map apps than me, so why not ask him what he uses? Google provides great value, I agree, but sometimes there's better stuff out there.
> essentially everyone in my social circle has moved on from going to Google first for information
As you said, you don't know how representative is your social circle. And you may underestimate your Google usage. ChatGPT often isn't a good substitute to Google. Looking at my history, a lot of my Google queries couldn't be answer by ChatGPT. Either because I need precise information, or recent information. Often I use Google just to redirect me to some specific site, typically wikipedia.
If you have a social circle where 100% of your contacts turn to an LLM first, not only have you buried yourself deep in a niche filter bubble, but you have also surrounded yourself with some of the planet's most misinformed people. If I were you, I'd be worried.
I don't think there's a measurable difference between his social circle and those who still rely on traditional media (news, TV, mainstream "journalism") to form their view of the world. It's just another cage.
Agreed. Reddit is a bubble too if you take note of the viewpoints there, but that reaches a wider variety of folks. Imagine how much of a bubble HN is.
HN is a bubble where people listen to what you say even if you argue against them, I haven't found any such place anywhere else before. I've found some people who listen even to people they disagree with on some niche forums, but never a community as big as this with density this high.
Other than listening and arguing in relatively good faith I don't think HN is a bubble, although such people probably have pretty different opinions than the typical person simply because they listen and change themselves more so it gets more refined.
Everyone wants their shares to go up, more growth, which can then lead to questionable actions on their part. Did you see the recent article about search enshittification (tldr; to make key metrics go up, make search worse so that users have to make more queries to get what they want)?
This might surprise people a lot.
In developing countries Google is synonymous with internet.
A lot of even young, people have a mental model of the internet as a subset of Google.
Seems like a situation where it’s worth it to be cognizant of your bubble.
I don’t know exactly what you meant by “our demographic”, but I’m a frequent reader of HN, work in tech, and generally stay up to date with all things tech, and yet… I don’t know if a single person that doesn’t still use Google as their go-to place for information. Before this comment I had never even heard of Perplexity or You.com. /shrug
It is probably massively better for niche technical searches which many technical people here do for their job, that is the Kagis userbase after all.
But such searches doesn't generate much ad revenue, so it doesn't do much to Google to lose them, all the people searching for articles about makeup or clothes or games or other things that has strong advertisement potential Google is awesome as a search engine, I haven't seen anyone say Kagi is better there. Kagi is only better for the searches that Google doesn't care about since they generate so little ad revue, Google uses so much compute per search that I'm not even sure such technical searches would be profitable for them to run, likely that would increase profits to lose them to Kagi.
> I guess our demographic is not predictive of the larger market.
I think it is, "our demographic" is just a few years ahead of the masses. For example MacBooks started getting popular with programmers way before they got popular with the masses.
funny you mentioned that because this demographic is the same demographic that predicted the demise of FB (it has grown in orders of magnitudes), hating on Tesla (also has been growing), didn't believe the viability of Dropbox idea.
When you say moved on can you share what you're using when you want/need to buy something online? Understand you moving on for information, but just curious what you're using these days for a specific product.
Only better quality alternative I found that has the same coverage and is available here.
FWIW for my purposes Marginalia is also a lot better quality (less annoying, more likely to give me the results I want - if it has them) than Google now but I cannot use it as my only search engine since the index is still small.
The other big ones (Bing, DDG) managed to still be worse than Google last time I tried despite Google goong out of their way to make it easy for others to walk past them.
There was another interesting one but they discriminated against non Americans and didn't even allow is to try so I don't know. It looked promising though, but as long as kagi don't betray us I get I stay with it.
I'm not a huge online shopper, but if the item isn't a total commodity that I just buy from Amazon without thinking, I generally ask Perplexity for suggestions and find the summaries/conclusions more helpful than clicking through a bunch of links delivered by Google.
I think Google still beats the various AI tools for any use case where you need to find a real world thing or decide between real world things. Which are also the primary types of queries that drive advertising revenue for Google. Yeah, ChatGPT has replaced searching Google for stack overflow answers, but it's not like you were clicking on ads when looking for programming answers anyway.
Perplexity especially is absolute garbage for 99% of people and their use cases. Google search is instant, available from every browser window and works with minimal keystrokes. LLM based search is such a bad experience on mobile where verifying the likely nonsense response is even more of an expensive task.
Just remember no one is auditing what views, likes and clicks count Google and Facebook tell you, you are getting. Advertisers just milk corporations. They dont care if the numbers are fake. They are now trained to tell everyone to spend more or you dont get attention someone else will.
As Goldharber once famously said , about the Attention Economy - people have limited attention to give anything but infinite capacity to receive attention.
No one likes to hear or believe they dont really have any influence when the system is signalling they do. So the ponzi scheme grows larger and larger.
There is a great book about it (from an ex-googler) called the Subprime Attention Crisis.
No one knows what to do about it so everyones head is buried deep in the sand.
We need new attention allocation systems that are not market driven.
Advertisers don't always insist on seeing evidence of genuine ROI but online advertising is more measurable than most other forms of ads and when the ROI slips a lot of ads customers notice and act accordingly.
Marketing is fascinating WRT to fraud because everyone benefits from it: the ad space provider, the ad marketplace, and the marketing team who needs to show performance.
I personally think this line of thought is due to people just not accepting how profitable advertising is to the end business. We're taking a clear and massive increase in profits that dwarf advertising costs. People don't advertise on meta/Google because they are duped by bots or some conspiracy between ads marketplace and marketing teams to hide the fraud. They advertise because their bottom line goes up and it's clear and direct. I once worked for a company where the only reason we'd ever drop some amount of advertising was due to supply constraints. We sold too much at certain points! It really does work.
Maybe not. But your brain noticed those brand names and when you need a product in that category they are more likely to come to mind.
It’s actually shocking if you think about products from your childhood which are no longer advertised and have completely dropped out of public consciousness.
Recently, I read on another HN thread that Kagi uses Google to build its index. If that's the case, the only way Kagi appears better than Google is that it doesn't show ads. Not sure, if my understanding is correct or not.
kagi allows you to downrank or even ban or boost certain websites so that when you serach, if there is a result from that website, it'll be downranked or boosted based on what you set up. It also automatically downranks websites that use shady ad trackers. They are slowly building their 'crawler' too but it's more nuanced regarding how they do this
Exactly and this is what they've been doing, per the recent trial and post on the guy who destroyed Search. That's what 90% market share let's you do, stuff another 1 or 2 ads before results, reword/suggest queries so they are more valuable, worsen results so people try more queries (and see more ads) for one search, articially boost the 2nd ad bid (in a second price auction) so that the winner has to pay more than 2nd place bid.
Been a shareholder a long time and 20%+ growth on 11-digit quarterly revenue still mindboggled me
Yes, and this strategy is only good for one or two pumps, because the advertisers will eventually notice the lower engagement rate per dollar of spend and reduce their commitment accordingly.
I'm curious what 2025 will look like for Elgoog. There are only so many desperate death throw spin tweaks available before the opportunities for indirection are exhausted.
I think so, as an early AdWords small advertiser, in the first cycle most of the inbound sales were explained by SEO and Ads, then SEO continued to be more important than Ads but Ads declined, then SEO was an impossible game and we run Ads just because we assigned a budget for Ads, then I hired three different agencies for optimizing AdWords campaigns for a niche market and nothing happened... so I started thinking AdWords is a kind of scam for small companies.
But Google and Meta can easily manufacture views for whoever pays the most by showing those ads more.
And then every other advertiser has compete with that price.
I mean if 2 big movies are being released this weekend and one producer is willing to spend 5 million on ads then pressure on the other producer goes up to spend 6. Its not that simple or scientific to connect how the movies perform to the ads. Same with politics. One jackass raises 100k to spend on ads then their opponent feels pressure to raise and spend the same or more. There is lot of fomo driving this machine.
On top of it Corps love to say the people are buying what they want. But google results show corps are spending a hell of a lot influence who buys what. So are people driving real demand or ad spend driving it?
Everyone is competing furiously for limited Attention. How do you get noticed in a room called the internet where everyone is trying to capture attention simultaneously. It feels like in such an environment of attention scarcity Google and Facebook are like the Oil Cartel. They can control the price.
As you said, The central point is where the attention goes, hence if you just prompt an LLM for a movie to watch now because it gives better answers the focus will be shifted against Google.
The layoffs and dividend are related -- the company's death rattle has been shaking a while now. Management is desperate to retain the confidence of the investor class.
Speaking as a shareholder here, a 2 trillion dollar valuation warrants extraordinary justification. Past performance is no guarantee of future results. As the AI models get better and better at compressing the world's information, people will be going directly to their model of choice
I completed the ML specialization for Georgia Tech's online CS MS program (https://omscs.gatech.edu/) a few year ago and found it tremendously valuable. I can't say that it changed the trajectory of my career, but I learned a lot, met some great people, and (in hindsight) had a ton of fun.
Keep in mind that most machine learning fundamentals haven't changed for decades. While new architectures/trends are always emerging, the lessons that you will take away from an academic program like OMSCS will be relevant for the rest of your career.
It's definitely a lot of work to build these systems (but also fun if you are competitive and like math/cs puzzles). While the margins are thin in stat arb, the absolute profits are not and risk-adjusted (but not time-adjusted) returns quite good.
I defined stat arb in the glossary at the end but probably should have included a more obvious link.
TL;DR: this is a pairs trading strategy that relies on the (very strong) statistical assumption that the price of a token on a centralized and decentralized exchange will converge.
My longer definition below:
Stat arbs: in finance, statistical arbitrage generally refers to any trade where a pair of assets should statistically move in a certain way. However, there are degrees of should. TradFi traders might reason that Meta and Google are both in the ads business, so if Meta is relatively expensive and Google is relatively cheap, they should short the former and long the latter. However, this is a weak argument. Perhaps Meta is just a better business or Google has structural problems. A stronger stat arb thesis is that Royal Dutch Shell used to be traded on both American and European exchanges. If the shares were trading at different prices on each, nearly risk-free profits are available to those who close the spread. This is what stat arb means in a crypto setting. AVAX may be trading at slightly different prices on Binance and on various blockchains.
Isn’t AVAX transferable between blockchains and Binance? If so, this isn’t “stat arb”, it’s just regular arbitrage. When the instruments are fungible, it doesn’t matter whether there’s a statistical relationship; your profit is guaranteed once you complete the trade, regardless of how the prices move later.
The Royal Dutch/Shell case is different because the two stocks were not actually fungible, so the trade would only have been profitable if the two prices eventually converged.
PS I enjoyed the article and it’s clear you’ve spent a lot of time thinking about this space. I just can’t resist chiming in when it comes to well-established terminology.
In theory, yes, the tokens should be transferable between the blockchain and the CEX, but this link breaks all the time and for various reasons (see the crypto problems section). Lots of things that “should” happen in regular, efficient markets don’t in crypto.
That said, call it whatever you want it. Most people in crypto call CEX/DEX stat arb to differentiate from true risk-free arbitrage, but I agree that people coming from a traditional trading perspective would call this pure arb.
https://www.ddmckinnon.com/2024/10/03/dans-weekly-ai-speech-...