Tag: programming

Taming Real-Time Logging: Lessons Learned From the Trenches – Chris Apple – ADCx SF

  • Lobby
  • Tag Archives: programming

Join Us For ADC23 - London - 13-15 November 2023
More Info: https://audio.dev/
@audiodevcon

Taming Real-Time Logging: Lessons Learned From the Trenches - Chris Apple - ADCx SF

As audio programmers, we know the cardinal rule: "Thou Shalt Not Use Unbounded System Calls in the Realtime Thread". However, it can be useful to have log statements for debugging, posting periodic analytics, or monitoring the status of a long-running audio thread.

By combining open source libraries, we can create a simple logger that is lock-free, allocation-free, and wait-free. This logger allows us to use printf/cout/spdlog even from our rendering thread. This talk will provide a practical look at the design and creation of a real-time logger and how to avoid potential time-unbounded booby traps.

Slides: https://data.audio.dev/talks/ADCxSF/designing-smart-algorithms/slides.pdf
_
Chris Apple

Chris Apple is the Lead Audio Software Engineer at Spatial Inc. He has spent the last 8 years specializing in immersive audio at Dolby and Roblox, including working on the first implementation of Dolby Atmos Music. He has contributed to the development of immersive audio in the Roblox Game Engine, the Dolby Atmos Music Panner, and the integration of Atmos into Apple Logic Pro X.

Recently, Chris has focused on streamlining the audio pipeline at Spatial. He is passionate about modern audio programming best practices when writing performant, real-time safe code.

Edited by Digital Medium Ltd - online.digital-medium.co.uk
_

Organized and produced by JUCE: https://juce.com/
_

Special thanks to the ADC Team:

Sophie Carus
Derek Heimlich
Andrew Kirk
Bobby Lombardi
Tom Poole
Ralph Richbourg
Jim Roper
Jonathan Roper

#audiodevcon #audiodev #logging

Filed under: UncategorizedTagged with: , , , ,

Designing Smart Algorithms with Traditional DSP vs. Machine Learning – Amit Shoham – ADCx SF

  • Lobby
  • Tag Archives: programming

Join Us For ADC23 - London - 13-15 November 2023
More Info: https://audio.dev/
@audiodevcon

Designing Smart Algorithms with Traditional DSP vs. Machine Learning - Amit Shoham - ADCx SF

Smart algorithms such as wakeword detection, tempo detection, song recognition, and many others have become an integral component of countless applications. These algorithms rely on a wide range of machine learning and/or traditional DSP techniques. While machine learning techniques are now solving previously impossible problems, algorithms designed with more traditional engineering techniques often require fewer system resources and are easier to deploy. In this talk we'll compare and contrast machine learning and traditional engineering approaches, and discuss fundamental principles that will help you determine what mix of techniques is best for your application.

Slides: https://data.audio.dev/talks/ADCxSF/designing-smart-algorithms/slides.pdf
_

Amit Shoham

Amit Shoham has led dual careers as a musician and engineer. His professional roles have included DSP engineer, house music producer, DJ, computer vision engineer, remixer, deep learning engineer, mastering engineer, and troublemaker. Amit is currently a senior systems architect and algorithms guru at Artiphon, where his deep expertise in algorithm design and optimization helps bring to life innovative new musical instruments.

Streamed & Edited by Digital Medium Ltd - online.digital-medium.co.uk
_

Organized and produced by JUCE: https://juce.com/
_

Special thanks to the ADC22 Team:

Sophie Carus
Derek Heimlich
Andrew Kirk
Bobby Lombardi
Tom Poole
Ralph Richbourg
Jim Roper
Jonathan Roper

#audiodevcon #audiodev #deeplearning

Filed under: UncategorizedTagged with: , , , ,