Sarcasm can be tricky even for humans – let alone a computer.

That's why researchers at the speech technology laboratory of the University of Groningen Decided to build an AI sarcasm detector that can understand tone of voice and convey those emotions through emoji embedded in transcribed text.

One of the researchers who worked on the project, Xiyuan Gao, presented the work Thursday as part of a joint meeting held by the Acoustical Society of America and the Canadian Acoustical Society at the Shaw Center in Ottawa.

Usually, sentiment analysis simply “focuses on the text,” according to Gao.

The new approach goes deeper into how people say things, not just what they say, which could help fields like AI-assisted healthcare. The study's findings could also mean better AI virtual assistants that can understand our tone.

Connected: These 'Expressive Avatar' Deepfakes from a Billion Dollar AI Startup Look Scary Real

The study took a multi-layered approach to sarcasm, assessing both what they could hear and what the speaker said on paper.

The researchers first evaluated the audio recordings based on pitch, speed of speech and other factors to understand the emotion beneath each word.

They then transcribed the audio recordings into text and tagged each segment of text with emojis that reflected the emotional intent behind the speech.

“Our approach leverages the combined strengths of auditory and textual information along with emoticons for a comprehensive analysis,” Gao DECLARING in a press release.

Connected: Employers say they want to hire candidates with AI skills, but employees are still using AI tools in the office

Looking ahead, the researchers want their algorithm to be able to understand more sarcastic expressions and gestures.

“In addition, we would like to include more languages,” Gao said.

Voice cloning and AI generation has been top of mind lately OpenAI, Google and other tech companies release more advanced AI models with more exciting voices than ever.

OpenAI appeared The sound engine last month, but stopped short of releasing the realistic text-to-speech voice generator due to “the potential for synthetic voice abuse.”

Connected: OpenAI is delaying the release of its new AI voice generator – Here's why

Other projects presented at the acoustic conference include spider webs on microphones and the ways to it reduce the noise in social settings.