HN Explorer

8 comments

[ 3.7 ms ] story [ 32.4 ms ] thread
maccam912
Where can I learn more about these phrases to say to capture as many sounds as possible? They listed two in the article, but now I'm curious how they decided which sentences to say, how efficient they are, if they account for the same sounds but with different qualities, like going up at the end if it's a question? It's a rabbit hole I need to find the entrance to.
rpearl
A good starting point is speech synthesis, particularly diphone synthesis
neckardt
This doesn’t answer all of your questions, but Apple has an accessibility feature which lets you generate a voice based on your own. It’ll prompt you to say various phrases, at the end you’ll have a synthetic voice to use. Info on it here: https://support.apple.com/en-us/104993
cwilkes
“Back in July 2025, six years before Apple introduced Siri, Susan Bennett made some recordings.”

The author is from the future?

cassepipe
God here. There's been a slight hiccup in the linearity of time. Our teams are working on it. Everything should be back to normal in a few hours ago.
gnabgib
GPTs everywhere:

- "fast forward" - prime indicator

- Yes.. voice actors who do this kind of work know this, it's in the contract

- Hallucinated date (2025, could be a typo for 2005 since Siri appeared in 2011[0])

- Bad-math: 2011 isn't 15 years ago, nor is 2025, nor is 2005

- RE: The last two back links: there's no incredible improvement, and any revolutionary new cameras would be Sony's

[0]: https://en.wikipedia.org/wiki/Siri

mmansoor78
Supercar Blondie is now a content farm too. Siphoning off other's content and LLMed.
locusofself
I'm convinced that "Sky" (one of the voice options for ChatGPT) is Rashida Jones, but ChatGPT won't cop to it.