- Experts have developed a device that can detect when someone is being sarcastic
- It works by examining the pitch, speaking rate and energy in speech
Our friends across the pond are known to struggle with British sarcasm from time to time.
But improved Anglo-American relations may be in the offing, now that experts have developed a device that can detect when someone is being sarcastic.
A team from the University of Groningen has created an algorithm that analyzes someone’s speech to find out whether he or she is using irony.
It works by examining the pitch, speaking rate and energy in speech and then converting the speech into text so that it can be further analyzed for language signals.
“We extracted acoustic parameters such as pitch, speaking rate and energy from speech and then used automatic speech recognition to convert the speech into text for sentiment analysis,” said author Xiyuan Gao.
Our friends across the pond are known to struggle with British sarcasm from time to time. In the photo: British comedian Frankie Boyle
‘We then assigned emoticons to each speech segment, reflecting its emotional content.
‘By integrating these multimodal signals into a machine learning algorithm, our approach leverages the combined powers of auditory and textual information along with emoticons for comprehensive analysis.’
Sarcasm – which Oscar Wilde once described as the lowest form of humor but the highest form of intelligence – is notoriously difficult to convey through text and can be easily misinterpreted even in person.
The team says they are optimistic about the performance of their algorithm, but are already looking for ways to improve it further.
Sarcasm – which Oscar Wilde once described as the lowest form of humor but the highest form of sarcasm – is notoriously difficult to convey through text and can easily be misinterpreted, even in person (stock image)
“There are a plethora of expressions and gestures that people use to emphasize sarcastic elements in their speech,” Ms Gao added.
‘These must be better integrated into our project. We would also like to add more languages and apply new sarcasm recognition techniques.’
Sentiment analysis, which focuses on text, is already being used to detect online hate speech and collect online customer opinions.
Meanwhile, emotion recognition based on speech can be applied to AI-assisted healthcare, the team said.