The new office mate’s prank will be to switch your AI’s voice to a Kardashian, Fran Drescher, Pauly Shore...
Plus I am sure that LLM engines could command a premium by shrewdly licensing such talent as James Earl Jones, Majel Barrett, or Milla Jovovich?
But it seems like the current trend is novel/generic voices in order to avoid suits/fees and pioneer new territory. Isn‘t Siri‘s personality a recognizable celebrity by now?
Because they think it makes them sound Smart? Because it's safer to fit in with the crowd than to Not? Because having a genuine personality is Difficult?
I would assume the opposite. People use it to sound dumb, not smart, so as to sound non-threatening, but also so as to sound non-assertive. It's used to defuse potential conflict or perceived disagreeableness. Generally the affect of a late-stage conflict-adverse institutionalism that punishes assertive or dominant behaviors. Uptalk is the "I'm showing my belly" of English affects.
The point of language is to have shared communication. That people adopt language standards isn't a moral failing, and accent is a legitimate part of it as it's more organic/instinctual to human interaction than grammar/diction classes.
As far as I can tell, it's an "I am not done talking yet" thing in many contexts. That's why people employing it usually drop the uptalk for their last sentence in a string of sentences (unless it's an actual question).
I'm having a rough couple of months, I'll be honest. [that should be read in the least sincere "my pleasure to serve you at the window" voice you can muster]
There’s no shortage of YouTube videos aware of the uptalk or upspeak annoyances but I found this one from 1994. It seems to have spread from California Valleygirl-speak and then nationwide to college campuses.
How does LLM training get influenced (weighted) by pop culture speaking styles?
I wish it were only a California thing. The Valley Girl uptalk/vocal fry thing seems to have spread across the country. Turn on the local news station in any region of the country and you'll hear it. Everyone is for some reason trying to sound like the Real Housewives Of Orange County.
NPR's even been full of it for more than a decade now. I think at some point (the '00s?) they really relaxed their elocution standards for hosts & reporters.
It definitely makes reporting feel trustworthy and serious? When almost every statement sounds like a tentative question?
Unless you train on twitch streamers. There seems to be an unspoken rule that any successful streamer has to move to LA. If you train on youtubers you get a surprising Mormon bias instead.
I would be surprised if the majority of the training data is licensed from the speakers.
Malaysian here. Uptalk is used more to turn a sentence into a suggestion. Something like "Hey, there's one dim sum left," to suggest that I'm taking this but you can challenge it. I could see why ChatGPT would adopt it. It's trying to be polite.
Often it's in a tonal particle, "One dim sum left meh." But it's possible in trying to artificially combine tone and text, the uptalk is moved up.
But the tell tones of a Malaysian accent is it's clipped. Instead of "I don't like that idea," it becomes "Don't like it." ChatGPT may be written American, so as an accent, it would sound closer to, "I- don't like, that idea."
And sentences often end in an elongated manner, "I wrote that is essay you wanted~". The elongated ends are quite common in many SEA accents as well, especially Thai.
I believe OpenAI wants ChaptGPT to have a tone that is more casual and less professional or uptight than it was before....
And so ChatGPT relies on the training data to know what that means so it leads to it talking like this as this is what the training data is filled with.
AI accents are incredibly good these days, especially eleven labs. ChatGPT is not a leader in this. I spent about $20 on this before just because I like the sound of its voice.
Readit News
Plus I am sure that LLM engines could command a premium by shrewdly licensing such talent as James Earl Jones, Majel Barrett, or Milla Jovovich?
But it seems like the current trend is novel/generic voices in order to avoid suits/fees and pioneer new territory. Isn‘t Siri‘s personality a recognizable celebrity by now?
I reckon it took off because of telephones.
https://youtu.be/z756L_CkakU
It definitely makes reporting feel trustworthy and serious? When almost every statement sounds like a tentative question?
I've gotten old enough to now wonder if my dialect sounds like something from another world and era to younger folks in my region.
The way I felt about most of the Hollywood actors I heard from before technicolor was the norm.
I would be surprised if the majority of the training data is licensed from the speakers.
Often it's in a tonal particle, "One dim sum left meh." But it's possible in trying to artificially combine tone and text, the uptalk is moved up.
But the tell tones of a Malaysian accent is it's clipped. Instead of "I don't like that idea," it becomes "Don't like it." ChatGPT may be written American, so as an accent, it would sound closer to, "I- don't like, that idea."
And sentences often end in an elongated manner, "I wrote that is essay you wanted~". The elongated ends are quite common in many SEA accents as well, especially Thai.
Deleted Comment
/s
And so ChatGPT relies on the training data to know what that means so it leads to it talking like this as this is what the training data is filled with.