Neural networks have become the architecture of choice for building Artificial Intelligence applications. They can reason about a large amount of data and discover the patterns within them. One particular set of neural network architectures focus on modeling time: recurrent neural networks. These are some of our favorite creative uses of recurrent neural networks. Each of these also has a link to the source code, if you want to learn how it’s done.
1. Samim Winiger, Obama RNN & TED Talk RNN
SEED: Democracy
“Good morning. One of the borders will be able to continue to be here today. We have to say that the partnership was a partnership with the American people and the street continually progress that is a process and distant lasting peace and support that they were supporting the work of concern in the world. They were in the streets and communities that could have to provide steps to the people of the United States and Afghanistan. In the streets — the final decade of the country that will include the people of the United States of America. Now, humanitarian crisis has already rightly achieved the first American future in the same financial crisis that they can find reason to invest in the world.
Thank you very much. God bless you. God bless you. Thank you.”
After creating a popular neural net modeling Obama’s speeches, Samim moved on to TED Talks. See both below.
2. Emily Daniels, Poetic Inner Join
they are sure
he is dreadful doing jane s
pensive emmanothing long ago
remember it all these
kindred irelandcordiality me
bowlinggreen pale behind
painful was thereno it attentions
lament is her air
excellent reproachunwell distinctly
good spirits at least humble
finer threeandtwentyshe felt only all
ran colonel light
shall i cared yoursalas true indeed
beautiful gracious
temporary ten
This series of “haikus” is built from a neural net that learned the works of Jane Austen.
3. David Ha, Sketch RNN
David Ha built this neural net to generate fake Kanji with TensorFlow.
4. Daniel Johnson, Composing Music with Recurrent Neural Networks
Daniel Johnston used a Classical midi piano database to teach a machine to write piano pieces. Head to his blog to check these out.
5. Keunwoo Choi, LSTMMetallica
Theoretically, Metallica doesn’t need a drummer anymore.
6. Ross Goodwin, NeuralSnap
There are a lot of image-captioning neural nets out there, but this is one of my favorites. Rather than just guessing what it’s looking at, it writes a poem about you.
The post 6 Open Source Creative Explorations Using Recurrent Neural Networks appeared first on Kadenze Blog.