Re:Neural audio codecs: how to get audio into LLMs
by tsol on Tuesday October 21, 2025 @11:00AM
(#45656667)
Did they respond differently depending on what race they thought you were? I'm surprised they would even do that honestly. I thought they were trained on text conversations which presumably wouldn't have any of that to learn from.
[ Parent
| on HN ]
Re:Neural audio codecs: how to get audio into LLMs
by OisinMoran on Tuesday October 21, 2025 @11:10AM
(#45656799)
You can often tell where someone is from from text alone! There are plenty of idiosyncrasies even in how different English speaking countries use the language.
Re:Neural audio codecs: how to get audio into LLMs
by fragmede on Tuesday October 21, 2025 @12:05PM
(#45657486)
Like, what do you mean? Are there, like, particular mannerisms that people from some regions that are hella unique to those regions?
Re:Neural audio codecs: how to get audio into LLMs
by robotresearcher on Tuesday October 21, 2025 @12:19PM
(#45657662)
I say old chap, what colour are your mummy’s wellies?
Re:Neural audio codecs: how to get audio into LLMs
by ctxc on Tuesday October 21, 2025 @12:20PM
(#45657679)
Clever!
Re:Neural audio codecs: how to get audio into LLMs
by ElevenLathe on Tuesday October 21, 2025 @03:48PM
(#45660775)
You betcha!
Re:Neural audio codecs: how to get audio into LLMs
by xwolfi on Tuesday October 21, 2025 @11:56PM
(#45664749)
All my Indian colleagues say "I agree with the same", this "the same" turn of phrase was so strange to me I had to ask (I'm French, so I have my own silly quirks, like I forget non-vocal plural(s<-- see, often I don't write that s)). They told me it was like that in Hindi so they just reproduce the pattern and it's grammatically acceptable.
For French people like me, false friends are immediately noticeable: for instance, "actually" to mean "now" instead of "in fact".
Re:Neural audio codecs: how to get audio into LLMs
by Xmd5a on Wednesday October 22, 2025 @10:07PM
(#45677405)
>like I forget non-vocal plural(s<-- see, often I don't write that s)).
c'est infernal. infernal
Re:Neural audio codecs: how to get audio into LLMs
by anotherhue on Tuesday October 21, 2025 @11:12AM
(#45656828)
Ah stop
Re:Neural audio codecs: how to get audio into LLMs
by j45 on Tuesday October 21, 2025 @12:05PM
(#45657478)
There are subtle differences in language where two groups can be speaking English and one is having a completely different conversation without saying much.
Re:Neural audio codecs: how to get audio into LLMs
by dotancohen on Tuesday October 21, 2025 @02:27PM
(#45659642)
This is quite the reason my wife evolved into my ex-wife.
Re:Neural audio codecs: how to get audio into LLMs
by thwarted on Tuesday October 21, 2025 @11:25AM
(#45656985)
If it did, it responded based on the accent it picked up on not race, because race and accent are orthogonal, correlation does not imply causation.
Re:Neural audio codecs: how to get audio into LLMs
by dotancohen on Tuesday October 21, 2025 @02:28PM
(#45659653)
Are denying that race and accent are highly correlated?
Re:Neural audio codecs: how to get audio into LLMs
by thwarted on Tuesday October 21, 2025 @07:52PM
(#45663273)
No, I'm saying that it is more meaningful to use what is directly derived rather than what is an indirect assumption. There is already issues with people erroneously considering whatever LLMs output as truth, the last thing anyone needs is an LLM claiming someone like Idris Elba is a white Briton because of his accent. We don't need automated phrenology machines, and that's what "determined your race from your voice" is pretty close to.
Re:Neural audio codecs: how to get audio into LLMs
by vessenes on Wednesday October 22, 2025 @12:00AM
(#45664768)
Pre-nerf the 4o voice model had a wide range of expressivity, and it would match affect (still tries to do this) and idiolect of listeners if asked. Nowadays there's a list of accents that are considered "hate-ish" and a list that aren't.
I will elide the rant inside me that west coast 20 somethings get to decide if speaking in a certain accent is racist or "bad". But it's a heartfelt rant.