Generation of Chords from Song Lyrics
Bisheshank C. Aryal: bcaryal Ryan Lee: xlee4 Daniel Kang: dkang33 Chai Harsha: charsha
In this project, we created a text-to-chord generation model that translates English text into a sequence of harmonically coherent chords using an encoder-decoder architecture. The model leverages GRU layers and embedding representations to learn relationships between text and chords. Our approach utilizes the Guitar Resource chord-lyric dataset, comprising 500 songs, to train the model. Despite challenges like limited annotated data and chord representation complexity, the model demonstrates promising results in producing chord progressions that align with the sentiment of the lyrics.
Built With
- python
- tensorflow
Log in or sign up for Devpost to join the conversation.