I don't usually use the “snark” tag or “/s” when writing sarcastic stuff, but that habit can sometimes get me in trouble. My opinion with the sarcastic tag is that it seems a bit like a crutch. My thinking is, “If it's a good sarcasm, people will 'get' it even without a tag warning of its presence.” If not, it's bad, bad timing, or just my fault for putting inappropriate sarcasm out there. ”
Indeed, in general, it requires special mental effort to make sure that the harassment is viable, for example, taking into account the norms of a particular forum. Unlike many interpersonal verbal situations (where sarcasm is signaled by emphasis, volume, emotion, and tone of voice), sarcasm in written discourse must be successful based on the forum, audience, and context. Because it's almost always an important consideration to take into account. You can't rely on visual or auditory cues. For example, it's clear that what is considered hateful on this site is not hateful on a site like Breitbart, and vice versa, even if the actual words are the same.
Tom Hawking, writing in Popular Science magazine, said:
Due to the nature of sarcasm, it can be difficult to identify just by looking at the words. Sarcastic statements often say one thing but mean another. This requires deriving the actual meaning of the statement from other, more subtle clues.
Teaching computers to understand and respond to sarcasm and sarcasm has long been understood as a challenge. But Stanley Kubrick has prepared us well for that eventuality.
Dave: Open the podbay door, Hal.
Hal: Sorry, Dave. Unfortunately that's not possible.
Dave: What's the problem?
Hal: I think you know as well as I do what the problem is.
***
Dave: Okay, Hal. Enter through the emergency airlock.
Hal: Without a space helmet, Dave, you'd think that would be pretty difficult.
(2001: A Space Odyssey (1968), Screenplay: stanley kubrick and Arthur C. Clarke, Arthur C. Clarke's uncredited story “The Sentinel'') (in this example, HAL does not actually know that Dave Bowman is merely “difficult” to re-enter the ship). He designed it that way (in his mind, anyway). The ironic robot “TARS” also appears in the movie directed by Christopher Nolan. Interstellar.
As The Guardian's science editor Ian Sample explained, researchers are now 'An AI-powered sarcasm detector helps you identify when you're at your lowest wit and highest level of intelligence. ”
In research presented Thursday at a joint meeting of the Acoustical Society of America and the Acoustical Society of Canada in Ottawa, Xiyuan Gao, a doctoral student in the lab, said the group has He described how he trained a neural network on emotional content. Clips from American sitcoms such as Friends and The Big Bang Theory. database, The file, known as Mustard, was compiled by researchers in the United States and Singapore, who labeled sentences from TV programs as sarcastic and built their own detectors.
One of the scenes the AI was trained on is: Leonard's wasted effort To escape from the locked room in The Big Bang Theory, Sheldon says, “It's just a privilege to be able to observe how your mind works.” In another episode of Friends, Ross invites Rachel to come over to his house and assemble furniture with Joey and Chandler, but Chandler commented, “Yes, I'm really looking forward to it.''
By training on text and audio, along with scores reflecting the emotional content of words spoken by actors, the AI was able to detect sarcasm in unlabeled interactions on sitcoms almost 75% of the time. Further research is being conducted in the lab using synthetic data to further improve accuracy, but the work is awaiting publication.
Hawking's article details how this “commentary” was actually done. This is based on a model proposed in a research paper “earlier” (2019), when AI was still making its way into the public consciousness.
a Short summary of the study Documents published on the conference site explain how the model works. Words are extracted from the audio data using automatic speech recognition and assigned emoticons to represent the underlying emotion. This emoji maps to multimodal cues such as tone of voice and broader conversational context. The authors suggest that their approach “leverages the strengths of each modality.” [and] compensate[s] It alleviates the limitations of pitch perception by providing complementary cues essential for accurate irony interpretation. ”
Another presentation at the same conference considered the role of pitch perception in sarcasm and identified specific acoustic features that tend to be present when a person is being sarcastic.
In particular, we focused on changes in F0 (fundamental frequency). lowest frequency of a particular person's voice.a specific change in this frequency In English, irony is often characterizedand Identify these changes Therefore, it has become a fairly reliable method of identifying ironic phrases.
The fact that AI was able to identify sarcasm in 75% of conversation snippets from comedy shows like Friends (as mentioned above) is actually pretty impressive. As one researcher pointed out, entering additional variables into the model, such as eyebrow movements or fake smiles, could improve that rate. More importantly, he points out that real people don't “get” sarcasm 100% of the time. In fact, far from it.
Like everything in the 21st century, the pace of development of AI snark recognition will increase exponentially in the coming years. Matt Koller, one of the researchers interviewed in the Sample Guardian article, wonders:
[W]The worst will happen if machines embrace our newfound skills and begin to cast irony on us. “If I ask, 'Do you have time to ask a question?' And it says, 'Sure,' I might think, do I or don't I?”
And since sarcasm always has an element of duplicity built into it, the AI may ultimately decide not to tell us that it's “just a joke.” This is often more interesting as the AI becomes aware of it.
Once this is released, I'm planning to eat sushi somewhere. Good night everyone. seriously!
