FluCoMa Podcast #05: Zakiya Leeming and Christopher Melen

The fifth episode of the FluCoMa podcast where Jacob Hart talks to PRiSM’s Zakiya Leeming and Christopher Melen.

The Podcast

Chapters

  1. Introduction (0:0)

  2. Zakiya & Christopher present themselves (1:52)

  3. What is PRiSM? (4:53)

  4. What is SampleRNN? (10:43)

  5. The premise for Sad Dog Eating (15:5)

  6. Clarinet multiphonics (24:12)

  7. How does Christopher approach his practice? (29:58)

  8. Working with SampleRNN (35:30)

  9. Composing with SampleRNN (40:20)

  10. The latent machine learning sound (52:32)

  11. Pairing this approach with FluCoMa for the engineer’s workflow… (57:3)

  12. …and for the composer’s workflow (1:2:12)

  13. SampleRNN content in Sad Dog Eating (1:8:58)

  14. Running SampleRNN at home (1:16:7)

  15. Approaches to data as an aesthetic object (1:20:21)

  16. Data and audio in all their forms (1:25:19)

Overview

In the 5th episode of the FluCoMa Podcast I talk with two people from Royal Northern College of Music’s PRiSM research centre (Practice and Research in Science and Music). I talk with composer Zakiya Leeming, who undertook her PhD at the centre, and was recently appointed the position of Artist and Producer in Residence. I also talk to Christopher Melen who has been the Research Software Engineer at PRiSM since 2019. In the podcast we discuss their relationship of composer and engineer, working on Zakiya’s piece Sad Dog Eating which made use of PRiSM’s implementation of SampleRNN, a deep-learning algorithm that learns from a corpus of audio to generate new content.

Here are the links to some of the things that were discussed during the podcast:

  • Dadabots.