h a l f b a k e r yThunk.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Say words question you take. Put give too much answer very very.
Advertisement same same good very I give. Not waste.
Too much benefit is good. I take much information inside. Then, insufficient good very. I come back have same same. Learn speech not necessary even construct. They five.
[link]
|
| |
I too five. We have so much in common. |
|
| |
baked on most instruction leaflets I come across. |
|
| |
So if I'm reading this idea right it's "gibberish is funny ha ha, so make a chatbot that speaks gibberish" Am I reading this idea right? |
|
| |
I don't think it's quite gibberish, or not necessarily at least. I'm pretty sure linguists can predict the type of errors and mispronunciations someone will make in a second language based on vocal and grammatical habits from their first language. Classic ones are mixing up genders when coming from a language that doesn't use gendered pronouns. Or confusing the sound of L and R when coming from a language that don't distinguish the sounds. |
|
| |
But of course not all mistakes are predictable, so some randomness (gibberish) should be sprinkled in. I knew a Chinese real estate agent in the US who somehow learned the word "sewage" incorrectly and would constantly refer to it as "sausage". I couldn't correct him without laughing so I just let it be. |
|
| |
All your base are belong to us. |
|
| |
Thanks [a1]. Better in Yodish, it sounds |
|
| |