- cross-posted to:
- news@lemmy.world
- cross-posted to:
- news@lemmy.world
Anybody know if it is hard to make a synthetic voice, just like you can make a synthetic face or synthetic music? Or is that the point: that a synthetic voice ended up sounding like her?
It’s just as easy. I was surprised to learn that they hired a voice actress. I guess hiring voice actors is cheaper than risking having to explain technology to a jury.
Similar voices, clearly different actresses. Imagine refusing a job then feeling entitled to some sort of compensation…
Here is an alternative Piped link(s):
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.
Honestly sounds nothing like Scarlett Johansson to me. OpenAI is shit for many reasons but this seems just another “AI” hysteria wave being pushed.
I wouldn’t say nothing like her, it’s uncomfortably close especially after being told no. But after the initial uproar, I was shocked how dissimilar it was when I actually heard it.
Maybe you people all have shitty headphones or ears but they sound nothing alike. This is the first time I heard them, and I really thought by all the drama that they were at least comparable. People need to clean their ears.
I grew up playing classical piano and saxophone. Moved into jazz saxophone and then jazz bass, and eventually played with a (terrible) rock band for a couple of years on bass. Technically (although I torture the term here a bit) I was a professional musician.
Not knowing anything else about you, it’s probably a safe bet to put my ear up against yours any day. And while I agree they are clearly different people, if you think they sound nothing alike the bet sounds even safer.
It’s funny that you think your stage damaged ears would be somehow “experienced” or whatever type of point you’re trying to make here. You blasted your ear drums and now have bad hearing. Either way, you fail to make out the distinctions between the voices clear enough to separate them, I don’t, because I have still excellent hearing unfortunately. Claiming you have better ears because you cannot hear the difference is nonsensical.
It’s funny that you think your stage damaged ears would be somehow “experienced” or whatever type of point you’re trying to make here.
I’ve spent my whole life training my ear by listening to and performing music with other people. . .and you can’t see the value of that when it comes to listening to stuff? Impressively ignorant.
In my first post (which you responded to) “I was shocked how dissimilar it was”
in my second post (which you responded to) “I agree they are clearly different people”
You in this post “you fail to make out the distinctions between the voices … you cannot hear the difference”
Man, you’re reading comprehension is even worse than your ear. lol
You also said they’re “uncomfortably close”. But nice cherry picking. Maybe just form less contradictionary opinions and understand what I was replying to, you know, reading comprehension as you said. And yes, if you’re a stage musician and handling loud as fuck instruments then your ears will be damaged by that. This isn’t even a controversial take but simply the reality of those things and you as an alleged musician should know this.
So they used AI to determine this? So I’m sure the result must be totally accurate.
What else does the article say? Hmmm let’s see. “The researchers found that Sky was also reminiscent of other Hollywood stars, including Anne Hathaway and Keri Russell. The analysis of Sky often rated Hathaway and Russell as being even more similar to the AI than Johansson.” Alright that proves it! Clearly this voice was based on Scarlett Johansson!
Many things are called “AI models” nowadays (unfortunately due to the hype). I wouldn’t dismiss the tools and methodology yet.
That said, the article (or the researchers) did a disservice to the analysis by not including a link to the report (and code) that outlines the methodology and how the distribution of similarities look. I couldn’t find a link in the article and a quick search didn’t turn up anything.
I am going to follow this shitshow with great interest
The bastard went ahead and did it anyway, even when told no.
“Our analysis shows that the two voices are similar but likely not identical,” Berisha said.
They also point out the main differences between the two voices in the paragraphs below this quote. I do believe that they hired a voice actress and that they didn’t train on SJ’s voice, or at least not entirely. However, I wouldn’t be surprised if there was big push for finding a voice similar to SJ’s voice in Her, no matter how much they deny this.
🙄👌👍
This is so dumb, it’s clearly not her.
Its quite obvious it isn’t her if you actually listen to a comparison video.
https://m.youtube.com/watch?v=aQ8UVSXnefk
It was quite gross seeing people actually defend this nonsense. Scarlet doesn’t own her whole spectrum of tone, nor do the directors of Her own the concept of a bubbly female assistant.
Tom Waits, IIRC, successfully sued the makers of a commercial for mimicking his voice, so there is precedent. And the fact that OpenAI reached out to her about using her voice, and that Altman tweeted the word “Her” on its own as a teaser of the product makes it pretty clear that they had knowledge and intent. I think she’s likely to have a pretty solid case here if she chooses to pursue it.
Here is an alternative Piped link(s):
https://m.piped.video/watch?v=aQ8UVSXnefk
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.
This is the best summary I could come up with:
Actress Scarlett Johansson’s voice bears a striking resemblance to OpenAI’s now-pulled “Sky” personal assistant, according to an artificial intelligence lab analysis conducted by researchers at Arizona State University.
The researchers measured Sky, based on audio from demos OpenAI delivered last week, against the voices of around 600 professional actresses.
The lab study shows that the voices of Sky and Johansson have undeniable commonalities – something many listeners believed, and that now can be supported by statistical evidence, according to Arizona State University computer scientist Visar Berisha, who led the voice analysis in the school’s College of Health Solutions and the College of Engineering.
Berisha said while the study analyzed a vast array of subtle vocal features, it also zoomed in on several particular dimensions of each voice and teased out some differences.
Altman himself amped up the speculation on the day of the release by posting on X one word, “her,” the name of the 2013 romantic sci-fi film in which a lonely man falls in love with a superintelligent computer operating system voiced by Johansson.
The company says he deferred to OpenAI Chief Technology Officer Mira Murati, who told NPR she didn’t even know what Johannson sounded like until people were comparing Sky to the actress.
The original article contains 786 words, the summary contains 207 words. Saved 74%. I’m a bot and I’m open source!