Creative Prediction Projects
Charles Martin - The Australian National University
Charles Martin. Lecturer in Computer Science, Australian
National University. [email protected]
Benedikte Wallace. PhD Researcher, University of Oslo.
Learning to Predict Sequences
Melody to Harmony in MicroJam
- Gain an overview of DL for music generation.
- Develop a melody to harmony sequence to sequence model
- Train the model on matched melody/harmony sequences
- Use MicroJam-sourced data as input and see if the generated harmonies make sense!
Seq-to-Seq Music Generation
- Understand the Transformer architecture.
- Implement your own Transformer (e.g., in Keras).
- Find a musical dataset that could be trained.
- Train your model, listen to the results and find a way to evaluate them.
Generating colour palettes from audio data
- Gain an overview of DL for audio processing.
- Obtain a dataset of audio and video (or colour) data.
- Try different neural network designs and evaluate the results. (Even a simple fully-connected ANN might work well!)
- Gain an overview of the main DL methods used for motion generation including RNNs, MDRNNs, and world models.
- Find a dataset of motion capture or other movement data (or capture one yourself!)
- Train the ANN and evaluate its generative abilities.