From Doja Cat’s hit single ‘Freak’ to Daði Freyr’s Eurovision classic ‘Think About Things’, a number of songs have found fame on TikTok in recent months.
The social media platform has surged in popularity this year, making it a prime target for singers hoping to see their songs go viral.
Now, researchers from Musiio have used artificial intelligence to develop what they claim is the ‘perfect TikTok song.’
Ysabel Wong, a researcher at Musiio, explained: “A ‘sound’ starts to go viral usually because a TikToker would shoot a short video with a repeatable format (like a dance or joke) so that others may create their own versions.
“What ends up happening is that a large bulk of the popular videos on TikTok are people doing the exact same thing, and over time these formats change as TikTok trends evolve.”
- 1 What we need to do to raise number of women in STEM careers
- 2 Ethics should guide use of artificial intelligence
- 3 Why ethics should guide artificial intelligence use
- 4 Youth hold the key to growth of information technology in Africa
To develop the song, the team analysed the top 120 songs on the TikTok Trending playlist in May 2020.
The analysis revealed that half of the top hits were hip-hop, while most were between either 60 to 80 ppm or 105 to 130 bpm.
Finally, the analysis revealed that the top songs tend to avoid negative moods, and instead are more likely to be powerful, energetic or quirky tracks.
Based on this analysis, Musiio’s AI got to work producing the perfect TikTok track.
The AI opted for a sample of gamelan - an Indonesian genre of instrumental music, alongside a distorted hip-hop bass drum.
Overall, the system gave the track a 41 per cent hip hop rating, and a mood that was 64 per cent quirky and 41 per cent powerful.
The resulting track is rather unusual but pretty catchy - although it’s yet to take off on TikTok!