Yeat Large Language Model

 _____      _        ____
| ____|_ __(_) ___  / ___| _ __   ___ _ __   ___ ___ _ __
|  _| | '__| |/ __| \___ \| '_ \ / _ \ '_ \ / __/ _ \ '__|
| |___| |  | | (__   ___) | |_) |  __/ | | | (_|  __/ |
|_____|_|  |_|\___| |____/| .__/ \___|_| |_|\___\___|_|
                          |_|

← back to index

Yeat Large Language Model

Fine tuning GPT-2 to talk like Yeat

repo: https://github.com/EricSpencer00/yeat-llm

Yeat LLM generates song lyrics from a model trained on all of Yeat's public lyrics from Genius. Sample output:


twizzy)
It's up
I got so high (Yeah, yeah)
I been going on snappin', yeah
Yeah (Yeah, yeah)
Up on the sky
Goin' down (Goin' out)
Up on the ceiling (Goin' up, goin')
Up off the coast (Goin' up, goin')
Goin' down, goin' down (Goin')
Goin' down (Goin')
Goin' down (Goin')
Goin' down, goin', goin', goin', goin', goin', goin', goin', goin' down (Goin', goin', goin', goin', goin', goin', goin')
Goin', goin', goin', goin', goin', goin', goin', goin', goin', goin', goin', goin

Set up the Python environment as described here.

Scraping needs a Genius API key. With the key in place, make requests via scrape_lyrics.py, which writes the lyrics into the /songs directory. Then train the model. The trained model is queried through yeat_bot.py.

Neither the original lyrics nor the trained models are public, for copyright reasons. The same pipeline builds a model for a different artist by changing which artist is scraped.