Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
(Vaguely) reminiscent of ChaCha’s text message service, circa 2008. A friend of mine was constantly using it.
 
Good but don't think such a service will ever be available outside the US. Wonder what OpenAI will announce in the remaining days.
 
  • Like
Reactions: mganu
It seemed obvious to me this isn’t about creating a user experience but is for something OpenAI could use instead.

So I fired up a search for “how to create a voice clone” and the results discussing Retrieval-based Voice Conversion or the RVC model show that between 10-20 minutes of high quality audio source is needed to get the most accurate clone.

Sounds like OpenAI wants a huge data sample of your voices. For what? Who knows.
 
It seemed obvious to me this isn’t about creating a user experience but is for something OpenAI could use instead.

So I fired up a search for “how to create a voice clone” and the results discussing Retrieval-based Voice Conversion or the RVC model show that between 10-20 minutes of high quality audio source is needed to get the most accurate clone.

Sounds like OpenAI wants a huge data sample of your voices. For what? Who knows.
To train their model. At the end of the day data is king, and if Apple has been atrocious with Siri and soon enough witb Apple Inteligence it’s because they don’t have the data.
 
I heard on the radio this morning, that ChatGPT and Meta have joined forces for a whiole, so you can access Chat GPT via WhatsApp. Text only in the UK.

They are trying to get more people to use ChatGPT as the majority of people are not that interested in the AI stuff.
I don't use Whats app andcertainl;y will not use ChatGPT.
 
Wait, people still make phone calls?
Admittedly this is an easier way to access ChatGPT by voice than by using Apple Intelligence, because you can initiate a call hands-off using Siri (and you can do it on an iPhone that doesn’t support Apple Intelligence). One would have thought that Apple would provide an easier way.
 
So I fired up a search for “how to create a voice clone” and the results discussing Retrieval-based Voice Conversion or the RVC model show that between 10-20 minutes of high quality audio source is needed to get the most accurate clone.
The phone isn’t a high-quality audio source though.
 
“Call in th next 15 minutes and you’ll get not 1 but 2 non-stick frying pans!"
 
With any reporting on things like this we need to know about any privacy issues, especially when something is "free" which means users are often the product. How is user voice data being used in these "free" calls? Is everyone who calls in to the "free" number using their voice for ClosedAI voice training? We need more rigorous and responsible reporting in this gilded age of robber barons 2.0.
 
1734626320886.png

“Call me now!”
 
  • Like
Reactions: meson
It seemed obvious to me this isn’t about creating a user experience but is for something OpenAI could use instead.

So I fired up a search for “how to create a voice clone” and the results discussing Retrieval-based Voice Conversion or the RVC model show that between 10-20 minutes of high quality audio source is needed to get the most accurate clone.

Sounds like OpenAI wants a huge data sample of your voices. For what? Who knows.
I immediately thought of voice cloning when I read the title. I don't think they need that much time just to record the voiceprint.
 
  • Like
Reactions: neotint
That trend of dialing letters instead of numbers is unknown here in Germany. Instead companies prefer to have a number with patterns that are easy to remember. The problem with the letters is that there is no letter on the "1". That drastically reduces the number of possible numbers. Over 52% of all seven digit numbers have at least a 1 in them.
 
  • Like
Reactions: neotint
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.