https://github.com/bendavidaaron/dob-bylan-songwriter
This is a generative neural network to emulate Bob Dylan's songwriting
https://github.com/bendavidaaron/dob-bylan-songwriter
Last synced: 3 months ago
JSON representation
This is a generative neural network to emulate Bob Dylan's songwriting
- Host: GitHub
- URL: https://github.com/bendavidaaron/dob-bylan-songwriter
- Owner: BenDavidAaron
- Created: 2018-03-25T02:24:05.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2018-04-10T20:13:41.000Z (about 7 years ago)
- Last Synced: 2025-01-18T03:28:38.098Z (5 months ago)
- Language: Python
- Homepage:
- Size: 31.5 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Virtual Bob Dylan
This is a recurent neural network to generate novel Bob Dylan lyrics.
This repository will be used for deigining a model, once implemented, I'll build a webapp to serve the model.
Thanks to [Blazking](https://github.com/Blaziking) for his [Ngram based Lyric Generator](https://github.com/Blaziking/Lyrics-generation-using-Ngrams), which contained a nicely cleaned and preprocessed set of Dylan lyrics. I've now copied it into my own repo, since I am not using any of his modeling or preprocessing codes, but I want to leave credit where credit is due.
So Far:
* I've tokenized the text
* Encoded the tokens
* Instantiated a model
* Trained a model for one epic on one batch
* Generated text resembling stuttering (`Hey Mr Tambourine Man Sha Sh sh shee shee`)
TODO:
* Rent some gpu time on GCP or AWS
* Tune the model training and hyperparamaters so it can approximate Dylan's Lyricism